Make your own free website on Tripod.com

Solaris 2.x - tuning your TCP/IP stack and more

 Source: http://www.rvs.uni-hannover.de/people/voeckler/tune

Table of contents

1. History and introduction
2. TCP connection initiation
3. Retransmission related parameters
4. Path MTU discovery
5. Further advice, hints and remarks
5.1 Common TCP timers
5.2 Erratic IPX behaviours
5.3 Common TCP parameters
6. Windows, buffers and watermarks
7. Tuning your system
7.1 Things to watch
7.2 General entries in the file /etc/system
7.3 100 Mbit ethernet related entries
7.4 How to find further entries
8. Recommended patches
9. Related books and software
10. Uncovered material
11. Startup scripts

1. History and introduction

This page and the related work have a long history in gathering. I started out peeking wide eyed over the shoulders of two people from a search engine provider when they were installing the German server of a customer of my former employer. My only alternative resource of tuning information was the brilliant book TCP/IP Illustrated 1 by Stevens. I started gathering all information about tuning I was able to get my hands upon. The cumulation of these you are experiencing on these pages.

 Solaris allows you to tune, tweak, set and reset various parameters related to the TCP/IP stack while the system is running. Back in the SunOS 4.x days, one had to change various C files in the kernel source tree, generate a new kernel, reboot the machine and try out the changes. The Solaris feature of changing the important parameters on the fly is very convenient.

 Many of the parameters I mention in the rest of the document you are reading are time intervals. All intervals are measured in milliseconds. Other parameters are usually bytecounts, but a few times different units of measurements are used and documented. A few items appear totally unrelated to TCP/IP, but due to the lack of a better framework, they materialized on this page.

 Most tunings can be achieved using the program ndd. Any user may execute this program to read the current settings, depending on the readability of the respective device files. But only the super user is allowed to execute ndd -set to change values. This makes sense considering the sensitive parameters you are tuning. Details on the use of ndd can be obtained from the respective manual page.

 

ndd /dev/tcp \?                  # show all parameter keys
ndd /dev/tcp tcp_mss_def         # show the value to this key
ndd -set /dev/ip ip_forwarding 0 # switch off forwarding
All keys starting out with ip_  have to be used with the pseudo device /dev/ip. Analog behaviour is true for the keys starting in tcp_  etc. Andres Kroonmaa kindly supplied a nifty script to check all existing values for a network component (tcp, udp, ip, icmp, etc.). Usually I do the same thing using Perl.

 

2. TCP connection initiation

This section is dedicated exclusively to the various queues and tunable variable(s) used during connection instantiation. The socket API maintains some control over the queues. But in order to tune anything, you have to understand how listen and accept interact with the queues. For details, see the various Stevens books mentioned in the literature section.

 When the server calls listen, the kernel moves the socket from the TCP state CLOSED into the state LISTEN, thus doing a passive open. All TCP servers work like this. Also, the kernel creates and initializes various data structures, among them the socket buffers and two queues:

incomplete connection queue
This queue contains an entry for every SYN that has arrived. BSD sources assing so_q0len entries to this queue. The server sends off the ACK of the client's SYN and the server side SYN. The connection get queued and the kernel now awaits the completion of the TCP three way handshake to open a connection. The socket is in the SYN_RCVD state. On the reception of the client's ACK to the server's SYN, the connection stays one round trip time (RTT) in this queue before the kernel moves the entry into the

 

completed connection queue
This queue contains an entry for each connection for which the three way handshake is completed. The socket is in the ESTABLISHED state. Each call to accept() removes the front entry of the queue. If there are no entries in the queue, the call to accept usually blocks. BSD source assign a length of so_qlen to this queue.

 

Both queues are limited regarding their number of entries. By calling listen(), the server is allowed to specify the size of the second queue for completed connections. If the server is for whatever reason unable to remove entries from the completed connection queue, the kernel is not supposed to queue any more connections. A timeout is associated with each received and queued SYN segment. If the server never receives an acknowledgement for a queued SYN segment, TCP state SYN_RCVD, the time will run out and the connection thrown away. The timeout is an important resistence against SYN flood attacks.

 

[connection queues]   [connection initiation]
Figure 1: Queues maintained for listening sockets.   Figure 2: TCP three way handshake, connection initiation.
Historically, the argument to the listen function specified the maximum number of entries for the sum of both queues. Many BSD derived implementations multiply the argument with a fudge factor of 3/2. Solaris <= 2.5.1 do not use the fudge factor, but adds 1, while Solaris 2.6 does use the fudge factor, though with a slightly different rounding mechanism than the one BSD uses. With a backlog argument of 14, Solaris 2.5.1 servers can queue 15 connections. Solaris 2.6 server can queue 22 connections.

 Stevens shows that the incomplete connection queue does need more entries for busy servers than the completed connection queue. The only reason for specifying a large backlog value is to enable the incomplete connection queue to grow as SYN arrive from clients. Stevens shows that moderately busy webserver has an empty completed connection queue during 99 % of the time, but the incomplete connection queue needed 15 or less entries in 98 % of the time! Just try to imaginge what this would mean for a really busy webcache like Squid.

 Data for an established connection which arrives before the connection is accept()ed, should be stored into the socket buffer. If the queues are full when a SYN arrived, it is dropped in the hope that the client will resend it, hopefully finding room in the queues then.

 According to Cockroft, there was only one listen queue for unpatched Solari <= 2.5.1. Solari >= 2.6 or an applied TCP patch 103582-12 or above splits the single queue in the two shown in figure 1. The system administrator is allowed to tweak and tune the various maxima of the queue or queues with Solaris. Depending on wether there are one or two queues, there are different sets of tweakable parameters.

 The old semantics contained just one tunable parameter tcp_conn_req_max which specified the maximum argument for the listen(). The patched versions and Solaris 2.6 replaced this parameter with the two new parameters tcp_conn_req_max_q0 and tcp_conn_req_max_q. A SunWorld article on 2.6 by Adrian Cockroft tells the following about the new parameters:

 

tcp_conn_req_max [is] replaced. This value is well-known as it normally needs to be increased for Web servers in older releases of Solaris 2. It no longer exists in Solaris 2.6, and patch 103582-12 adds this feature to Solaris 2.5.1. The change is part of a fix that prevents denial of service from SYN flood attacks. There are now two separate queues of partially complete connections instead of one.

 tcp_conn_req_max_q0 is the maximum number of connections with handshake incomplete. A SYN flood attack could only affect this queue, and a special algorithm makes sure that valid connections can still get through.

 tcp_conn_req_max_q is the maximum number of completed connections waiting to return from an accept call as soon as the right process gets some CPU time.

In other words, the first specifies the size of the incomplete connection queue while the second parameters assigns the maximum length of the completed connection queue. All three parameters are covered below.

 You can determine if you need to tweak this set of parameters by watching the output of netstat -sP tcp. Look for the value of tcpListenDrop, if available on your version of Solaris. Older versions don't have this counter. Any value showing up might indicate something wrong with your server, but then, killing a busy server (like squid) shuts down its listening socket, and might increase this counter (and others). If you get many drops, you might need to increase the appropriate parameter. Since connections can also be dropped, because listen() specifies a too small argument, you have to be careful interpreting the counter value. On old versions, a SYN flood attack might also increase this counter.

 Newer or patched versions of Solaris, with both queues available, will also have the additional counters tcpListenDropQ0 and tcpHalfOpenDrop. Now the original counter tcpListenDrop counts only connections dropped from the completed connection queue, and the counter ending in Q0 the drops from the incomplete connection queue. Killing a busy server application might increase either or both counters. If the tcpHalfOpenDrop shows up values, your server was likely to be the victim of a SYN flood. The counter is only incremented for dropping noxious connection attempts. I have no idea, if those will also show up in the Q0 counter, too.

 

tcp_conn_req_max
default 8 (max. 32), since 2.5 32 (max. 1024), recommended 128 <= x <= 1024

since 2.6 or 2.5.1 with patches 103630-09 and 103582-12 or above applied:
see tcp_conn_req_max_q and tcp_conn_req_max_q0

 The current parameter describes the maximum number of pending connection requests queued for a listening endpoint in the completed connection queue. The queue can only save the specified finite number of requests. If a queue overflows, nothing is sent back. The client will time out and (hopefully) retransmit.

 The size of the completed connection queue does not influence the maximum number of simultaneous established connections after they were accepted nor does it have any influence on the maximum number of clients a server can serve. With Solaris, the maximum number of file descriptors is the limiting factor for simultaneous connections, which just happened to coincide with the maximum backlog queue size.

 From the viewpoint of TCP those connections placed in the completed connection queue are in the TCP state ESTABLISHED, even though the application has not reaped the connection with a call to accept. That is the number limited by the size of the queue, which you tune with this parameter. If the application, for some reason, does not release entries from the queue by calling accept, the queue might overflow, and the connection is dropped. The client's TCP will hopefully retransmit, and might find a place in the queue.

 Solaris offers the possibility to place connections into the backlog queue as soon as the first SYN arrives, called eager listening. The three way handshake will be completed as soon as the application accept()s the connection. The use of eager listening is not recommended for production systems.

 Solari < 2.5 have a maximum queue length of 32 pending connections. The length of the completed connection queue can also be used to decrease the load on an overloaded server: If the queue is completely filled, remote clients will be denied further connections. Sometimes this will lead to a connection timed out error message.

 Naively, I assumed that a very huge length might lead to a long service time on a loaded server. Stevens showed that the incomplete connection queue needs much more attention than the completed connection queue. But with tcp_conn_req_max you have no option to tweak that particular length.

 When tuning tcp_conn_req_max, always do it with regards to the values of rlim_fd_max and rlim_fd_cur. This is just a rule of thumb. Setting your listen backlog queue larger than the number of filedescriptors available to you won't do you any good if your service time is long. A server shouldn't accept any further connections, if it has run out of descriptors. Even though new connection won't be thrown away with a long backlog, a server might want to reduce the size to as many connections as can be serviced simultaneously. Again, you have to consider your average service time, too.

 

There is a trick to overcome the hardcoded limit of 1024 with a patch. SunSolve shows this trick in connection with SYN flood attacks. A greatly increased listen backlog queue may offer some small increased protection against this vulnerability. On this topic also look at the tcp_ip_abort_cinterval parameter. Better, use the mentioned TCP patches, and increase the q0 length.

 

echo "tcp_param_arr+14/W 0t10240" | adb -kw /dev/ksyms /dev/mem
This patch is only effective on the currently active kernel, limiting its extend to the next boot. Usually you want to append the line above on the startup script /etd/init.d/inetinit. The shown patch increases hard limit auf the listen backlog queue to 10240. Only after applying this patch you may use values above 1024 for the tcp_conn_req_max parameter.
A further warning: Changes to the value of tcp_conn_req_max parameter in a running system will not take effect until each listening application is restarted. The backlog queue length is evaluated whenever an application calls listen(3N), usually once during startup. Sending a HUP signal may or may not work; personally I prefer to TERM the application and restart them manually or, even better, use a startup script.

 

tcp_conn_req_max_q0
since 2.5.1 with patches 103630-09 and 103582-12 or above applied: default 1024;

since 2.6: default 1024, recommended 1024 <= x <= 10240

 After installing the mentioned TCP patches, alternatively after installing Solaris 2.6, the parameter tcp_conn_req_max is no longer available. In its stead the new parameters tcp_conn_req_max_q and tcp_conn_req_max_q0 emerged. tcp_conn_req_max_q0 is the maximum number of connections with handshake incomplete, basically the length of the incomplete connection queue.

 In other words, the connections in this queue are just being instantiated. A SYN was just received from the client, thus the connection is in the TCP SYN_RCVD state. The connection cannot be accept()ed until the handshake is complete, even if the eager listening is active.

 To protect against SYN flooding, you can increase this parameter. Also refer to the parameter tcp_conn_req_max_q above. I believe that changes won't take effect unless the applications are restarted.

 

tcp_conn_req_max_q
since 2.5.1 with patches 103630-09 and 103582-12 or above applied: default 128;

since 2.6: default 128, recommended 128 <= x <= tcp_conn_req_max_q0

 After installing the mentioned TCP patches, alternatively after installing Solaris 2.6, the parameter tcp_conn_req_max is no longer available. In its stead the new parameters tcp_conn_req_max_q and tcp_conn_req_max_q0 emerged. tcp_conn_req_max_q is the length of the completed connection queue.

 In other words, connections in this queue of length tcp_conn_req_max_q have completed the three way handshake of a TCP open. The connection is in the state ESTABLISHED. Connections in this queue have not been accept()ed by the server process (yet).

 Also refer to the parameter tcp_conn_req_max_q0. Remember that changes won't take effect unless the applications are restarted.

 

tcp_conn_req_min
Since 2.6: default 1, recommened: don't touch

 This parameter specifies the minimum number of available connections in the completed connection queue for select() or poll() to return "readable" for a listening (server) socket descriptor.

 Programmers should note that Stevens describes a timing problem, if the connection is RST between the select() or poll() call and the subsequent accept() call. If the listening socket is blocking, the default for sockets, it will block in accept() until a valid connection is received. While this seems no tragedy with a webserver or cache receiving several connection requests per second, the application is not free to do other things in the meantime, which might constitute a problem.

3. Retransmission related parameters

The retransmission timeout values used by Solaris are way too agressive for wide area networks, although they can be considered appropriate for local area networks. SUN thus did not follow the suggestions mentioned in RFC 1122. Newer releases of the Solaris kernel are correcting the values in question:

 

The recommended upper and lower bounds on the RTO are known to be inadequate on large internets. The lower bound SHOULD be measured in fractions of a second (to accommodate high speed LANs) and the upper bound should be 2*MSL, i.e., 240 seconds.
Besides the retransmit timeout (RTO) value two further parameters R1 and R2 may be of interest. These don't seem to be tunable via any Solaris' offered interface that I know of.

 

The value of R1 SHOULD correspond to at least 3 retransmissions, at the current RTO. The value of R2 SHOULD correspond to at least 100 seconds.

 [...]

 However, the values of R1 and R2 may be different for SYN and data segments. In particular, R2 for a SYN segment MUST be set large enough to provide retransmission of the segment for at least 3 minutes. The application can close the connection (i.e., give up on the open attempt) sooner, of course.

Great many internet servers which are running Solaris do retransmit segments unnecessarily often. The current condition of European networks indicate that a connection to the US may take up to 2 seconds. All parameters mentioned in the first part of this section relate to each other!

 As a starter take this little example. Consider a picture, size 1440 byte, LZW compressed, which is to be transferred over a serial linkup with 14400 bps and using a MTU of 1500. In the ideal case only one PDU gets transmitted. The ACK segment can only be sent after the complete PDU is received. The transmission takes about 1 second. These values seem low, but they are meant as 'food for thought'. Now consider something going awry...

 [New] Solaris 2.5.1 is behaving strange, if the initial SYN segment from the host doing the active open is lost. The initial SYN gets retransmitted only after a period of 4 * tcp_rexmit_interval_initial plus a constant C. The time is 12 seconds with the default settings. More information is being prepared on the retransmission test page.

 The initial lost SYN may or may not be of importance in your environment. For instance, if you are connected via ATM SVCs, the initial PDU might initiate a logical connection (ATM works point to point) in less than 0.3 seconds, but will still be lost in the process. It is rather annoying for a user of 2.5.1 to wait 12 seconds until something happens.

 

tcp_rexmit_interval_initial
default 500, since 2.5.1 3000, recommended >= 2000 (500 for special purposes)

 This interval is waited before the last data sent is retransmitted due to a missing acknowledgement. Mind that this interval is used only for the first retransmission. The more international your server is, the larger you should chose this interval.

 Special laboratory environments working in LAN-only environments might be better off with 500 ms or even less. If you are doing measurements involving TCP (which is almost always a bad idea), you should consider lowering this parameter.

 

tcp_rexmit_interval_min
default 200, recommended >= 1000 (200 for special purposes)

 After the initial retransmission further retransmissions will start after the tcp_rexmit_interval_min interval. BSD ususally specifies 1500 milliseconds. This interval should be tuned to the value of tcp_rexmit_interval_initial, e.g. some value between 50 % up to 200 %. The parameter has no effect on retransmissions during an active open, see my accompanying document on retransmissions.

 

The tcp_rexmit_interval_min doesn't display any influence on connection establishment with Solaris 2.5.1. It does with 2.6, though. The influence on regular data retransmissions, or FIN retransmissions I have yet to research.
tcp_ip_abort_interval
default 120000, since 2.5 480000, recommended 600000

 This interval specifies how long retransmissions for a connection in the ESTABLISHED state should be tried before a RESET segment is sent. BSD systems default to 9 minutes.

 

tcp_ip_abort_cinterval
default 240000, since 2.5 180000, recommended ?

 This interval specifies how long retransmissions for a remote host are repeated until the RESET segment is sent. The difference to the tcp_ip_abort_interval parameter is that this connection is about to be established - it has not yet reached the state ESTABLISHED. This value is interesting considering SYN flood attacks on your server. Proxy server are doubly handicapped because of their Janus behaviour (like a server towards the downstream cache, like a client towards the upstream server).

 According to Stevens this interval is connected to the active open, e.g. the connect(3N) call. But according to SunSolve the interval has an impetus on both directions. A remote client can refuse to acknowledge an opening connection up to this interval. After the interval a RESET is sent. The other way around works out, too. If the three-way handshake to open a connection is not finished within this interval, the RESET Segment will be sent. This can only happen, if the final ACK went astray, which is a difficult test case to simulate.

 To improve your SYN flood resistence, SUN suggests to use an interval as small as 10000 milliseconds. This value has only been tested for the "fast" networks of SUN. The more international your connection is, the slower it will be, and the more time you should grant in this interval. Proxy server should never lower this value (and should let Squid terminate the connection). Webservers are usually not affected, as they seldom actively open connections beyond the LAN.

 

tcp_rexmit_interval_max
default 60000, RFC 1122 recommends 240000 (2MSL), recommended 1...2 * tcp_close_wait_interval

since 2.6: default 240000

 All previously mentioned retransmissions related interval use an exponential backoff algorithm. The wait interval between two consequitive retransmissions for the same PDU is doubled starting with the mimimum.

 The tcp_rexmit_interval_max interval specifies the maximum wait interval between two retransmissions. If changing this value, you should also give the abort interval an inspection. The maximum wait interval should only be reached shortly before the abort interval timer expires. Additionally, you should coordinate your interval with the value of tcp_close_wait_interval.

 

tcp_deferred_ack_interval
default 50, BSD 200, recommended 200 or 500

 This parameter specifies the timeout before sending a delayed ACK. The value should not be increased above 500, as required by RFC 1122. This value is of great interest for interactive services. A small number will increase the "responsiveness" of a remote service (telnet, X11), while a larger value can decrease the number of segments exchanged.

 The parameter might also interest to HTTP servers which transmit small amounts of data after a very short retrieval time. With a heavy-duty servers or in laboratory banging environment, you might encounter service times answering a request which are well above 50 ms. An increase to 500 might lead to less PDUs transferred over the network, because TCP is able to merge the ACK with data. Increases beyond 500 should not even be considered.

 Please note that Solaris recognizes the initial data phase of a connection. An initial ACK (not SYN) is not delayed. Therefore a request for a webservice (both, server or proxy) which does not fit into a single PDU can be transmitted faster. Web benchmarks will show this as improved performance. Also check the tcp_slow_start_initial Parameter.

 

tcp_deferred_acks_max
Since 2.6: default 8, recommended ?

 This parameter has something to do with the number of delayed acknowlegdements or the number of byte to be collected. My guess is that this parameter specifies the number of outstanding ACKs in interactive transfer mode. In this case tiny amounts of data are flowing in both direction. In contrast to my prior statement, you need not give this parameter a look when tuning bulk transfers, because its impact is on interactive transfers.

 

Good values for retransmission tuning don't beam into existence from a white source. Rather you should carefully plan an experiment to get decent values. Intervals from another site do not carry on without change to another Solaris system. But they might give you an idea where to start when chosing your own values.

 The next part looks at a few parameters having to do with retransmissions, as well.

 

tcp_slow_start_initial
since 2.5.1 with patch 103582-15 applied: default 1

since 2.6: default 1, recommended 2 for web services

 This parameter provides the slow-start bug discovered in BSD and Windows TCP/IP implementations for Solaris. More information on the topic can be found on the servers of SUN and in [Stevens III].

 To summarize the effect, a server starts sending two PDUs at once without waiting for an ACK. This is due to the ACK from connection initiation being counted as data ACK - compare with figure 2. The client immediately acknowledges both PDUs, thus undermining network congestion avoidance algorithms. The slow start algorithm does not allow this behaviour, compare with RFC 2001.

 Setting the parameter to 2 allows a Solaris machine to behave like it has the slow start bug, too. Well, IETF is said to make amends to the slow start algorithm, and the bug is now actively turned into a feature. SUN also warns:

It's still conceivable, although rare, that on a configuration that supports many clients on very slow-links, the change might induce more network congestions. Therefore the change of tcp_slow_start_initial should be made with caution.
[...] Future Solaris releases are likely to default to 2.
tcp_dupack_fast_retransmit
default 3

 Something to do with the number of duplicates ACKs. If we do fast retransmit and fast recovery algorithms, this many ACKs must be retransmitted until we assume that a segment has really been lost. A simple reodering of segments usually causes no more than two duplicate ACKs.

 

4. path MTU discovery

Whenever a connection is about to be established, the three-way handshake open negotiation, the segment size used will be set to the minimum of (a) the smallest MTU of an outgoing interface, and (b) from MSS announced by the peer. If the remote peer does not announce a MSS, usually the value 536 will be assumed. If path MTU discovery is active, all outgoing PDUs have the IP option DF (don't fragment) set.

 If the ICMP error message fragmentation needed is received, a router on the way to the destination needed to fragment the PDU, but was not allowed to do so. Therefore the router discarded the PDU and did send back the ICMP error. Newer router implementations enclose the needed MSS in the error message. If the needed MSS is not included, the correct MSS must be determined by trial and error algorithm.

 Due to the internet being a packet switching network, the route a PDU travels along a TCP virtual circuit may change with time. For this reason RFC 1191 recommends to rediscover the path MTU of an active connection after 10 minutes. Improvements of the route can only be noticed by repeated rediscoveries. Unfortunately, Solaris aggressively tries to rediscover the path MTU every 30 seconds. While this is o.k. for LAN environments, it is a grossly impolite behaviour in WANs. Since routes may not change that often, aggressive repetitions of path MTU discoveries leads to unnecessary consumption of channel capacity and elongated service times.

 Path MTU discovery is a far reaching and controversial topic when discussing it with local ISPs. But think, the discovery is at the foundation of IPv6. The PSC tuning page argues pro path MTU discovery, especially if you maintain a high-speed or long-delay (e.g. satellite) link.

 The recommendation I can give you is not to use the defaults of Solaris < 2.5. Please use path MTU discovery, but tune your system RFC conformant. You may alternatively want to switch off the path MTU discovery all together, though there are few situations where this is necessary.

 I was made aware of the fact that in certain circumstances bridges connecting data link layers of differing MTU sizes defeat pMTU discovery. I have to put some more investigation into this matter. If a frame with maximum MTU size is to be transported into the network with the smaller MTU size, it is truncated silently. A bridge does not know anything about the upper protocol levels: A bridge neither fragments IP nor sends an ICMP error.

 There may be work-arounds, and the tcp_mss_def is one of them. Setting all interfaces to the minimum shared MTU might help, at the cost of losing performance on the larger MTU network. Using what RFC 1122 calls an IP gateway is a possible, yet expensive solution.

 

ip_ire_pathmtu_interval
default 30000, since 2.5 600000, recommended 600000

 This timer determines the interval Solaris rediscovers the path MTU. An extremely large value will only evaluate the path MTU once at connection establishment.

 

ip_path_mtu_discovery
default 1, recommended 1

 This parameter switches path MTU discovery on or off. If you enter a 0 here, Solaris will never try to set the DF bit in the IP option - unless your application explicitly requests it.

 

tcp_ignore_path_mtu
default 0, recommended 0

 This is a debug switch! When activated, this switch will have the IP or TCP layer ignore all ICMP error messages fragmentation needed. By this, you will achieve the opposite of what you intended.

 

tcp_mss_def
default 536, recommended >= 536

 This parameter determines the default MSS (maximum segment size) for non-local destination. For path MTU discovery to work effectively, this value can be set to the MTU of the most-used outgoing interface descreased by 20 byte IP header and 20 byte TCP header - if and only if the value is bigger than 536.

 

5. Further advice, hints and remarks

This section covers a variety of topics, starting with various TCP timers which do not relate to previously mentioned issues. The next subsection throws a quick glance at some erratic behaviour. The final section looks at a variety of parameters which deal with the reservation of resources.

 

5.1 Common TCP timers

The current subsection covers three important TCP timers. First I will have a look at the keepalive timer. The timer is rather controversial, and some Solari implement them incorrectly. The next parameter limits the twice maximum segment lifetime (2MSL) value, which is connected to the time a socket spends in the TCP state TIME_WAIT. The final entry looks at the time spend in the TCP state FIN_WAIT_2.

 

tcp_keepalive_interval
default 7200000, recommended 0 <= x <= oo

 This value is one of the most controversial ones when talking with other people about appropriate values. The interval specified with this key must expire before a keep-alive probe can be sent. Keep-alive probes are described in the host requirements RFC 1122: If a host choses to implement keep-alive probes, it must enable the application to switch them on or off for a connection, and keep-alive probes must be switched off by default.

 Keep-alives can terminate a perfectly good connection (as far as TCP/IP is concerned), cost your money and use up transmission capacity (commonly called bandwidth, which is, actually, something completely different). Determining wether a peer is alive should be a task of the application and thus kept on the application layer. Only if you run into the danger of keeping a server in the ESTABLISHED state forever, and thus using up precious server resources, you should switch on keep-alive probes.

 

[Webserver response]
 Figure 3: A typical handshake during a transaction.

Figure 3 shows the typical handshake during a HTTP connection. It is of no importance for the argumentation if the server is threaded, preforked or just plain forked. Webservers work transaction oriented as is shown in the following simplified description - the numbers do not relate to the figure:

 

  1. The client (browser) initiates a connection (active open).
  2. The client forwards its query (request).
  3. The server (daemon) answers (response).
  4. The server terminates the connection (active close).
Common implementations need to exchange 9..10 TCP segments per HTTP connection. The keep-alive option as a HTTP/1.0 protocol and extensions can be regarded as a hack. Persistent connections are a different matter, and not shown here. Most people still use HTTP/1.0, especially the Squid users.

 The keep-alive timer becomes significant for webservers, if in step 1 the client crashed or terminates without the server knowing about it. This condition can be forced sometimes by quickly pressing the stop button of netscape or the Logo of Mosaic. Thus the keep-alive probes do make sense for webservers. HTTP Proxies look like a server to the browser, but look like a client to the server they are querying. Due to their server like interface, the conditions for webservers are true for proxies, as well.

 With an implementation of keep-alive probes working correctly, a very small value can make sense when trying to improve webservers. In this case you have to make sure that the probes stop after a finite time, if a peer does not answer. Solari <= 2.5 have a bug and send keep-alive probes forever. They seem to want to elict some response, like a RST or some ICMP error message from an intermediate router, but never counted on the destination simply being down. Is this fixed with 2.5.1? Is there a patch available against this misbehaviour? I don't know, maybe you can help me.

 I am quite sure that this bug is fixed in 2.6 and that it is safe to use a small value like ten minutes. Squid users should synchronize their cache configuration accordingly. There are some Squid timeouts dealing with an idle connection.

 

tcp_close_wait_interval
default 240000 (according to RFC 1122, 2MSL), recommended 60000, possibly lower

 As Stevens repeatedly states in his books, the TIME_WAIT state is your friend. You should not desperately try to avoid it, rather try to understand it. The maximum segment lifetime(MSL) is the maximum interval a TCP segment may life in the net. Thus waiting twice this interval ensures that there are no leftover segments coming to haunt you. This is what the 2MSL is about. Afterwards it is safe to reuse the socket resource.

 The parameter specifies the 2MSL according to the four minute limit specified in RFC 1122. With the knowledge about current network topologies and the strategies to reserve ephemerical ports you should consider a shorter interval. The shorter the interval, the faster precious resources like ephemerical ports are available again.

 A toplevel search engine implementor recommends a value of 1000 millisecond to its customers. Personally I believe this is too low for regular server. A loaded search engine is a different matter alltogether, but now you see where some people start tweaking their systems. I rather tend to use a multiple of the tcp_rexmit_interval_initial interval. The current value of tcp_rexmit_interval_max should also be considered in this case - even though retransmissions are unconnected to the 2MSL time. A good starting point might be the double RTT to a very remote system (e.g. Australia for European sites). Alternatively a German commercial provider of my aquaintance uses 30000, the smallest interval recommended by BSD.

 

tcp_fin_wait_2_flush_interval
BSD 675000, default 675000, recommended 67500 (one zero less)

 This values seems to describe the (BSD) timer interval which prohibits a connection to stay in the FIN_WAIT_2 state forever. FIN_WAIT_2 is reached, if a connection closes actively. The FIN is acknowledged, but the FIN from the passive side didn't arrive yet - and maybe never will.

 Usually webservers and proxies actively close connections - as long as you don't use persistent connection and even those are closed from time to time. Apart from that HTTP/1.0 compliant server and proxies close connections after each transaction. A crashed or misbehaving browser may cause a server to use up a precious resource for a long time.

 You should consider decreasing this interval, if netstat -f inet shows many connections in the state FIN_WAIT_2. The timer is only used, if the connection is really idle. Mind that after a TCP half close a simplex data transmission is still available towards the actively closing end. TCP half closes are not yet supported by Squid, though many web servers do support them (certain HTTP drafts suggest an independent use of TCP connections). Nevertheless, as long as the client sends data after the server actively half closed an established connection the timer is not active.

 

Sometimes, a Squid running on Solaris (2.5.1) confuses the system utterly. A great number of connection to a varying degree are in CLOSE_WAIT for reasons beyond me. During this phase the proxy is virtually unreachable for HTTP requests though, abnoxiously, it still answers ICP requests. Although lowering the value for tcp_close_wait_interval is only fixing symptoms indirectly, not the cause, it may help overcoming those periods of erratic behaviour faster than the default. The thing needed would be some means to influence the CLOSE_WAIT interval directly.

 

5.2 Erratic IPX behaviour

I noticed that Solari < 2.6 behave erratically under some conditions, if the IPX ethernet MTU of 1500 is used. Maybe there is an error in the frame assembly algorithm. If you limit yourself to the IEEE 802.3 MTU of 1492 byte, the problem does not seem to appear. A sample startup script with link in /etc/rc2.d can be used to change the MTU of ethernet interfaces after their initialization. Remember to set the MTU for every virtual interface, too!

 

Note, with a patched Solaris 2.5.1 or Solaris 2.6, the problem does not seem to appear. Limiting your MTU to non-standard might introduce problems with truncated PDUs in certain (admittedly very special) environments. Thus you may want to refrain from using the above mentioned script (always called second script in this document).
Additionally, I strongly suggest the use of a file /etc/init.d/your-tune (always called first script) which changes the tunable parameters. /etc/rcS.d/S31your-tune is a hardlink to this file. The script will be executed during bootup when the system is in single user mode. A killscript is not necessary. The section about startup scripts below reiterates this topic in greater depth.

 

5.3 Common TCP parameters

The following parameters have little impact on performance, nevertheless I reckon them worth noting here:

 

ip_forward_src_routed
default 1, recommended 0

 This parameter determines if IP datagrams can be forwarded which have the source routing option activated. The parameter has little meaning for performance but is rather of security relevance. Solaris may forward such datagrams, if the host route option is activated, bypassing certain security construct - possibly undermining your firewall. Thus you should disable it always, unless the host functions as a regular router (and no other services).

 

ip_icmp_err_interval
default 500, recommended 0

 Who of you does not know the strange behaviour of the second asterisk when doing a traceroute onto or via a Solaris host. If you set this value to exactly 0, traceroute will not give your host away as running Solaris. I am afraid I don't have any idea what kind of ghosts you invite by setting this parameter to 0. So far, it didn't hurt the hosts I used it upon.

 

udp_smallest_anon_port
tcp_smallest_anon_port
default 32768, recommended 8192

 This value has the same size for UDP and TCP. Solaris allocates ephemerical ports above 32768. Busy servers or hosts using a large 2MSL, see tcp_close_wait_interval, may want to lower this limit to 8192. This yields more precious resources, especially for proxy servers.

 A contra-indication may be servers and services running on well known ports above 8192. This parameter should be set very early during system bootup, especially before the portmapper is started.

 

udp_largest_anon_port
default 65535, recommended: see text

 This paramters has to be seen in combination with udp_smallest_anon_port. The traceroute program tries to reach a random UDP port above 32768 - or rather tries not to reach such a port - in order to provoke an ICMP error message from the host.

 Paranoid system administrator may want to lower the value for this reason down to 32767, after the corresponding value for udp_smallest_anon_port has been lowered. On the other hand, datagram application protocols should be able to cope with foreign protocol datagrams.

 If Squid or other UDP hyper-active applications are used, the lowering of this value can not be recommended. The respective TCP parameter tcp_largest_anon_port does not suffer this problem.

 

6. Windows, buffers and watermarks

This section is about windows, buffers and watermarks. It is still work in progress. The explanations available to me were very confusing (sigh), though the new Stevens helped to clear up a few things. If you have corrections to this section, please let me know and contribute to an update of the page. Many readers will thank you!

 

[buffers and fragmentation]
 Figure 4: buffers and related issues

Here just a short trip through the network layer in order to explain what happens where. Your application is able to send almost any size of data to the transport layer. The transport layer is either UDP or TCP. The socket buffers are implemented on the transport layer. Depending on your choice of transport protocol, different actions are taken on this level.

 

TCP
All application data is copied into the socket buffer. If there is insufficient size, the application will be put to sleep. From the socket buffer, TCP will create segments. No chunk exceeds the MSS.

 Only when the data was acknowledged from the peer instance, the data can be removed from the socket buffer! For slow connections or a slowly working peer, this implies a very long time some data uses up the buffer.

 

UDP
The socket buffer size of UDP is simply the maximum size of datagram UDP is able to transmit. Larger datagrams ought to elict the EMSGSIZE error response from the socket layer. With UDP implementing an unreliable service, there is no need to keep the datagram in the socket buffer.

 Please assume that there is not really a socket buffer for sending UDP. This really depends on the operating systems, but many systems copy the user data to some kernel storage area, whereas others try to eliminate all copy operations for the sake of performance.

 

The IP layer needs to fragment chunks which are too large. Among the reasons TCP prechunks its segments is the need to avoid fragmentation. IP searches the routing tables for the appropriate interface in order to determine the fragment size and interface.

 If the output queue of the datalink layer interface is full, the datagram will be discarded and an error will be returned to IP and back to the transport layer. If the transport protocol was TCP, TCP will try to resend the segment at a later time. UDP should return the ENOBUFS error, but some implementations don't.

 To determine the MTU sizes, use the ifconfig -a command. The MTUs are needed for some calculation to be done later in this section. With IPv4 you can determine the MSS from the interface MTU by substracting 20 Bytes for the TCP header and 20 Bytes for the IP header. Keep this in mind, as the calculation will be repeatedly necessary in the text following below.

 

$ ifconfig -a
lo0: flags=849 mtu 8232
        inet 127.0.0.1 netmask ff000000 
el0: flags=863 mtu 1500
        inet 130.75.215.xxx netmask ffffff00 broadcast 130.75.215.255
        ether xx:xx:xx:xx:xx:xx
hme0: flags=863 mtu 1500
        inet 130.75.5.xxx netmask ffffff00 broadcast 130.75.5.255
qaa0: flags=863 mtu 9180
        inet 130.75.214.xxx netmask ffffff00 broadcast 130.75.214.255
        ether xx:xx:xx:xx:xx:xx
fa0: flags=842 mtu 9188
        inet 0.0.0.0 netmask 0 
        ether xx:xx:xx:xx:xx:xx
I removed the uninteresing things. hme0 is the regular 100 Mbps ethernet interface. The 10 Mbps ethernet interface is called le0. el0 is the ATM LAN emulation (lane) interface. qaa0 is the ATM classical IP (clip) interface. fa0 is the interface that supports Fore's proprietary implementation of native ATM. Fore is the vendor of the installed ATM card. AFAIK you can use this interface to build PVCs or, if you are also using Fore switches, SVCs. You see an unconfigured interface there.

 The buffer sizes for sending and receiving TCP segment and for UDP datagrams can be tuned with Solaris. With the help of the netstat command you can obtain an output similar but unlike the following one. The data was obtained on a server which runs a Squid with five dnsserver children. Since the interprocess communcation is accomplished via localhost sockets, you see both, the client side and the server side of each dnsserver child socket.

 

$ netstat -f inet

 TCP
   Local Address        Remote Address    Swind Send-Q Rwind Recv-Q  State
-------------------- -------------------- ----- ------ ----- ------ -------
blau-clip.ssh        challenger-clip.1023 57344     19 63980      0 ESTABLISHED
localhost.38437      localhost.38436      57344      0 57344      0 ESTABLISHED
localhost.38436      localhost.38437      57344      0 57344      0 ESTABLISHED
localhost.38439      localhost.38438      57344      0 57344      0 ESTABLISHED
localhost.38438      localhost.38439      57344      0 57344      0 ESTABLISHED
localhost.38441      localhost.38440      57344      0 57344      0 ESTABLISHED
localhost.38440      localhost.38441      57344      0 57344      0 ESTABLISHED
localhost.38443      localhost.38442      57344      0 57344      0 ESTABLISHED
localhost.38442      localhost.38443      57344      0 57344      0 ESTABLISHED
localhost.38445      localhost.38444      57344      0 57344      0 ESTABLISHED
localhost.38444      localhost.38445      57344      0 57344      0 ESTABLISHED
The columns titled with Swind and Rwind contain values for the size of the respective send- and reception windows, based on the free space available in the receive buffer at each peer. The Swind column contains the offered window size as reported by the remote peer. The Rwind column displays the advertised window size being transmitted to the remote peer.

 An application can change the size of the the socket layer buffers with calls to setsockopt with the parameter SO_SNDBUF or SO_RCVBUF. Windows and buffers are not interchangable. Just remember: The buffers have a fixed size - unless you use setsockopt to change. Windows on the other hand depend on the free space available in the input buffer. The minimum and maximum requirements for buffer sizes are tuneable watermarks.

 

[buffers, watermarks and windows]
 Figure 5: buffers, watermarks and window sizes.

Figure 5 shows the relation of the different buffers, windows and watermarks. I decided to let the send buffer grow from the maximum towards zero, which is just a way of showing things, and does probably not represent the real implementation. I left out the different socket options as the picture is confusing enough.

 

Squid users should note the following behaviour seen with Solaris 2.6. The default socket buffer sizes which are detected during configuration phase are representative of the values for tcp_recv_hiwat, udp_recv_hiwat, tcp_xmit_hiwat and tcp_xmit_hiwat. Also note that enabling the hit object feature still limits hit object size to 16384 byte, regardless of what your system is able to achieve.

 

output from Squid 1.1.19 configuration script on a Solaris 2.6 host with the previously mentioned parameters all set to 64000. Please mind that these parameters do not constitute optimal sizes in most environments:
checking Default UDP send buffer size... 64000
checking Default UDP receive buffer size... 64000
checking Default TCP send buffer size... 64000
checking Default TCP receive buffer size... 64000
Buffers and windows are very important if you link via satellite. Due to the daterate possible but the extreme high round-trip delays of a satellite link, you will need very large TCP windows and possibly the TCP timestamp option. Only RFC 1323 conformant systems will achieve these ends. In other words, get a Solaris 2.6. For 2.5 systems, RFC 1323 compliance can be purchased as a Sun Consulting Special.

 Window sizes are important for maximum throughput calculations, too. As Stevens shows, you cannot go faster than the window size offered by your peer, divided by the round-trip time (RTT). The lower your RTT, the faster you can transmit. The larger your window, the faster you can transmit. If you intend to employ maximum window sizes, you might want to give tcp_deferred_acks_max another look.

 The network research laboratory of the German research network did measurements on satellite links. The RTT for a 10 Mbps link (if I remember correctly) was about 500 ms. A regular system was able to transmit 600 kbps whereas a RFC 1323 conformant system was able to transmit about 7 Mbps. Only bulk data transfer will do that for you.

 

 (1)   10 Mbps * 0.5 s = 5 Mbit = 625 KB
 (2)   512 KB = 4 Mbit = 0.5 s * 8 Mbps
 (3)   64 KB / 0.5 s = 128 KBps = 1 Mbps
The bandwidth-delay-product can be used to estimate the initial value when tweaking buffer sizes. The buffers then represent the capacity of the link. If we apply the bandwidth-delay-product calculations to the satellite link above, we get the following results: Equation 1 estimates the buffer sizes necessary to fully fill the 10 Mbps link. Equation 2 assumes that the buffer sizes were set to 512 KB, which would yield 8 Mbps. Slight deviation in the experiment may have been caused by retransmissions. Finally, equation 3 estimates the maximum datarate we can use on the satellite link, if limited to 64 KB buffers, e.g. Solaris <= 2.5.1. The 1 Mbps constitute an upper limit, as can be seen by the measured 600 Kbps.

 Squid users beware: As long as Squid does not implement HTTP/1.1 persistent connections, you will not get any decent HTTP transmissions via satellite. The average cached object sizes about 13 kbyte, thus you almost never get past the TCP slow start. While this may or may not be a big deal with terrestrial links, but you will never be able to fill a satellite pipe to a satisfactorily degree. Doing things in parallel might help. Only when reaching TCP congestion avoidance you will see any filling of the pipe.

 

tcp_cwnd_max
default 32768, since 2.? 65535, recommended 65535 for Solaris <= 2.5.1

since 2.6 262144 (finally!), no recommendations

 This parameter describes the maximum size the congestion window can be opened. The congestion window is opened as large as possible with any Solaris up to 2.5.1. A change to this value is only necessary for older Solaris systems, which defaulted to 32768. The Solaris 2.6 default looks reasonable, but you might need to increase this further for satellite links.

 Though window sizes beyond 64k are possible, mind that the window scale option is only announced during connection creation and your maximum windows size is 1 GByte (1,073,725,440 Byte). Also, the window scale option is only employed during the connection, if both sides support it.

 

tcp_recv_hiwat
default 8192, recommended 16384 (see text), Cockroft 32768, maximum 65535

Solaris 2.6 LFN bulk data transfer 131071 or above (see text)

 This parameter determines the maximum size of the initial TCP reception buffer. The specified value will be rounded up to the next multiple of the MSS. From the free space within the buffer the advertised window size is determined. That is, the size of the reception window advertised to the remote peer. Squid users will be interested in this value with regards to the socket buffer size the Squid auto configuration program finds.

 The previous table shows an Rwind value of 63980 = 7 * 9140. 9140 is the MSS of the ATM classical IP interface (clip) in host blau. The interface itself uses a MTU of 9180. For the standard builtin 10 Mbps or 100 Mbps IPX ethernet, you get a MTU of 1500 on the outgoing interface, which yields an MSS of 1460. The value of 57344 in the next Rwind line points to the lo0 (loopback) interface, MTU 8232, MSS 8192 and 57344 = 7 * 8192.

 Starting with Solaris 2.6 values above 65535 are possible, see the window scale option from RFC 1323. Only if the peer host also implements RFC 1323, you will benefit from buffer sizes above 65535. If one host does not implement the window scale option, the window is still limited to 64K. The option is only activated, if buffer sizes above 64K are used.

 For HTTP, I don't see the need to increase the buffer above 64k. Imagine servicing 1024 simultaneous connections. If both the TCP high watermarks of your system are tuned to 64k and your application uses the system's defaults, you would need 128M just for your TCP buffers!

 Squid's configuration option tcp_recv_bufsize lets you select a TCP receive buffer size, but if set to 0 (default) the kernel value will be taken, which is configurable with the tcp_recv_hiwat parameter. A buffer size of 16K is large enough to cover over 70 % of all received webobjects on our caches.

 

tcp_recv_hiwat_minmss
default 4, no recommendations

 This parameter influences the minimum size of the input buffer. The reception buffer is at least as large as this value multiplied by the MSS. The real value is the maximum of tcp_recv_hiwat round up to the next MSS and tcp_recv_hiwat_minmss multiplied by the MSS, in other words, something akin to:

  hiwat_tmp ~= ceil( tcp_recv_hiwat / MSS )
  real_size := MAX( hiwat_tmp, tcp_recv_hiwat_minmss ) * MSS
That way, however bad you misconfigure the buffers, there is a guaranteed space for tcp_recv_hiwat_minmss full segments in your input buffer.

 

udp_recv_hiwat
default 8192, recommended 16384 (see text), maximum 65535

 The highwater mark for the UDP reception buffer size. This value may be of interest for Squid proxies which use ICP extensively. Please read the explanations for tcp_recv_hiwat. Squid users will want at least 16384, especially if you are planning on using the hit object feature of Squid.

 If you see many dead parent detections in your cache.log file without cause, you might want to increase the receive buffer. In most environments an increase to 64000 will have a neglegible effect on the memory consumption, as most application, including Squid, use only one or very few UDP sockets, and only in an iterative way.

 Remember if you don't set your socket buffer explicitely with a call to setsockopt(), your default reception buffer will have about this size. Arriving Datagrams of a larger size might be truncated or completely rejected. Some systems don't even notify your receiving appliction.

 

tcp_xmit_hiwat
default 8192, recommended 16384 (see text), Cockroft 32768, maximum 65535

Solaris 2.6 LFN bulk data transfer 131071 or above (see text)

 This parameter influence a heuristic which determines the size of the initial send window. The actual value will be rounded up to the next multiple of the MSS, e.g. 8760 = 6 * 1460. Also do read the section on tcp_recv_hiwat.

 The table further to the top shows a Swind of 57344 = 7 * 8192. For the standard builtin 10 Mbps or 100 Mbps IPX ethernet, you get an MTU of 1500 on the outgoing interface, which yields a MSS of 1460.

 Starting with Solaris 2.6 values above 65535 are possible, see the window scale option from RFC 1323. Only if the peer host also implements RFC 1323, you will benefit from buffer sizes above 65535. If one host does not implement the window scale option, the window is still limited to 64K.

 I don't see the need to increase the buffer above 32K for HTTP applications. Imagine servicing 1024 simultaneous connections. If both TCP high watermarks of your system are tuned to 32K, you would need 64M just for your TCP buffers! Squid 1.1.x does not seem to use the socket option SO_SNDBUF to limit this memory hunger during runtime. Mind that the send buffer has to keep a copy of all unacknowledged segments. Therefore it is affordable to give it a greater size than the receive buffer. Again, 16K covers over 70 % of all transferred webdocuments on our caches, and 32K should cover 90 %.

 

udp_xmit_hiwat
default 8192, recommended 16384, maximum 65535

This refers to the highwater mark for send buffers. May be of interest for proxies using ICP extensively. Please refer to the explanations for tcp_xmit_hiwat. Squid users will want at least 16384, especially if you are planning on using the hit object feature of Squid.

 Please remember that there exists no real send buffer for UDP on the socket layer. Thus, trying to send a larger amount of data than udp_xmit_hiwat will truncate the excess, unless the SO_SNDBUF socket option was used to extend the buffer.

 

tcp_xmit_lowat
default 2048, no recommendations

 The current paramenter refers to the amount of data which must be available in the TCP socket sendbuffer until select or poll return writable for the connected file descriptor.

 Usually there is no need to tune this parameter. Applications can use the socket option SO_SNDLOWAT to change this parameter on a process local basis.

 

udp_xmit_lowat
default 1024, no recommendations

 The current paramenter refers to the amount of data which must be available until select or poll return writable for the connected file descriptor. Since UDP does not need to keep datagrams and thus needs no outgoing socket buffer, the socket will always be writable as long as the socket sendbuffer size value is greater than the low watermark. Thus it does not really make much sense to wait for a datagramm socket to become writable unless you constantly adjust the sendbuffer size.

 Usually there is no need to tune this parameter, especially not on a system-wide basis.

 

tcp_max_buf
default 262144, minimum 65536, no immediate recommendations

since 2.6 1048576, minimum 65536, no immediate recommendations
udp_max_buf
default 262144 (since 2.5), minimum 65536, no immediate recommendations

 Ok, I am still unsure about the application of this value, and how it fits the big picture (e.g. figure 5). Stevens laconically states that these values declare the maximum size of the send or receive buffer. Is this the sum of both, or for each? And why use 256K, if before Solaris 2.6 no such buffer sizes were possible? Or is that the unlikely sum of all buffers for a process? Until I investigate further into this parameter, here is my old, probably wrong, explanation, which reads:

 

This value refers to the default maximum socket buffer size. The value ought be adjusted according to the settings of the previous parameters. For many paths, this is not enough, and must be increased. Without RFC 1323 large windows, the application is not allowed to buffer more than 64 kB in the network, which is inadequate for almost all high speed paths. Yes, one can use the network as a buffer. You can calculate the capacity of this buffer by multiplying the bandwidth of your path with the roundtrip time (RTT). The product is commonly called the bandwidth-delay-product. Using many connections like a Squid with possibly an enlarged socket buffersize (see tcp_xmit_hiwat and tcp_recv_hiwat), you might want to increase the TCP value. 
[New] A few odd remarks at this point, concerning the recommendations given for the transmission buffer sizes. I decreased the recommendations of Adrain Cockroft in favour of a more conservative memory consumption. Also, with an average HTTP object size of 13 KByte, you can expect to fit over 50 % of all objects into the transmission buffer. On the other hand, larger objects which are to be transmitted by a cache or webserver may suffer in certain circumstances. Furthermore, I should recommend a generic transmission buffer size which is double the reception buffer size. This recommendation bases on the fact that unacknowledged segments occupy the send buffer until they are acknowledged.

 

7. Tuning your system

7.1 Things to watch

Did you reserve enough swap space? You should have at least as much swap as you have main memory. If you have little main memory, even double your swap. Do not be fooled by the result of the vmstat command - read the manpage and realize that the small value for free memory shown there is (usually) correct.

 With Solaris there seems to exist a difference between virtually generated processes and real processes. The latter is extremely dependend on the amount of virtual memory. To test the amount of both kinds of processes, try a small program of mine. Do start it at the console, without X and not as priviledged user. The first value is the hardlimit of processes, and the second value the amount of processes you can really create given your virtual memory configuration. Tweaking your ulimit values may or may not help.

 

7.2 General entries in the file /etc/system

The file /etc/system contains various very important resource configurable parameters for your system. You use these tunings to give a heavily loaded system more resources of a certain kind. Unfortunately a reboot is necessary after changing anything. Though one could schedule reboots after midnight, I advice against it. You should always check if your changes have the desired effect, and won't tear down the system.

 Adrian Cockroft severly warns against transporting an /etc/system from one system onto another, even worse, onto another hardware platform:

 

Clean out your /etc/system when you upgrade.
The most frequent changes are limited to the number of file descriptors, because the socket API uses filedescriptors for handling internet connectivity. You may want to look at the hardlimit of filehandles available to you. Proxies like Squid have to count twice for each connection: open request descriptors and either an open file or an open forwarding request descriptors.

 You are able to influence the tuning with the reserved word set. Use a whitespace to seperate the key from the keyword. Use an equals sign to separate the value from its key. There are a few examples in the comments of the file.

 Please, before you start, make a backup copy of your initial /etc/system. The backup should be located on your root filesystem. Thus, if some parameters fail, you can always supply the alternative, original system file on the boot prompt. The following shows two typically entered parameters:

 

* these are the defaults of the system
set rlim_fd_max=1024
set rlim_fd_cur=64
WARNING! SUN does not make any guarantees for the correct working of your system, if you use more filedescriptors than 4096. Personally, my old fvwm window manager did quit working alltogether. In my case, I compiled it on a Solaris 2.3 or 2.4 system and transferred it always onwards to a 2.5 system. After compiling the fvwm95, it worked to my satisfaction.

 If you experience SEGV core dumps from your select(3c) system call after increasing your file descriptors above 4096, you have to recompile the affected programs. Especially the select(3c) call is known to the Squid users for its bad tempers concerning the maximum number of file descriptors. SUN remarks to this topic:

 

The default value for FD_SETSIZE (currently 1024) is larger than the default limit on the number of open files. In order to accommodate programs that may use a larger number of open files with select(), it is possible to increase this size within a program by providing a larger definition of FD_SETSIZE before the inclusion of <sys/types.h>.
Note: This does not work as expected. See text below.
I did test this suggestion by SUN, and a friend of mine tried it with Squid Caches. The result was a complete success or diseaster both times, depending on your point of view: If you can live with supplying naked women to your customers instead of bouncing company logos, go ahead and try it. If you really need to access filedescriptors above 1024, don't use select(), use poll() instead! poll() is supposed to be faster with Solaris, anyway. A different source mentions that the redefinition workaround mentioned above works satisfactorily; not for me, neither with Squid.

 At the pages of VJ are a some tricks which I incorporated into this paper, too. Personally I am of the opinion that the VJ pages are not as up to date as they could be.

 Many parameters of interest can be determined using the sysdef -i command. Please keep in mind that many values are in hexadecimal notation without the 0x prefix. Another very good program to see your system's configuration is sysinfo, the program. Refer to the manpages how to invoke this program.

 

rlim_fd_cur
default 64, recommended >= 1024

 This parameters defines the softlimit of open files you can have. The currently active softlimit can be determined from a shell with something like

ulimit -Sn
Use at your own risk values above 1024, especially if you are running old binaries. A value of 4096 may look harmless enough, but may still break old binaries.

 Another source mentions that using more than 8192 filedescriptors is discouragable. It mentions that you ought to use more processes, if you need more than 4096 file descriptors. On the other hand, an ISP of my acquaintance is using 16384 descriptors to his satisfaction.

 The predicate rlim_fd_cur <= rlim_fd_max must be fullfilled.

 

rlim_fd_max
default 1024, recommended >=4096

 This parameter defines the hardlimit of open files you can have. For a Squid and most other servers, regardless of TCP or UDP, the number of open filedescriptors per user process is among the most important parameter. The number of filedescriptors is one limit on the number of connections you can have in parallel. You can find out the value of your hardlimit on a shell with something like

ulimit -Hn
You should consider a value of at least 2 * tcp_conn_req_max and you should provide at least 2 * rlim_fd_cur. The predicate rlim_fd_cur <= rlim_fd_max must be fullfilled.

 Use at your own risk values above 1024. SUN does not make any warranty for the workability of your system, if you increase this above 1024. Squid users of busy proxies will have to increase this value, though. A good starting seems to be 16384 <= x <= 32768. Remember to change the Makefile for Squid to use poll() instead of select(). Also remember that each call of configure will change the Makefile back, if you didn't change Makefile.in.

 

maxusers
default 249 ~= Megs RAM (Ultra-2/2 CPUs/256 MB), min 8, max 2048, no recommendations

 This parameter determines the size of certain kernel data structures which are initialized at startup. There is strong indication that the default is determined from the main memory in megs. It might also be a function of the available memory and/or architecture.

 The defaults of the parameters max_nprocs, maxuprc, ufs_ninode, ncsize and ndquot will be determined from this parameter's value. The greater you chose the number for maxusers, the greater the number of the mentioned resources. The relation in strictly proportional: A doubling of maxusers will (more or less) double the other resources.

 Adrian Cockroft advises against a setting of maxusers. The kernel uses a lot of space while keeping track of the RAM usages within the system, therefore it might need to be reduced on systems with gigabytes of main memory.

 

max_nprocs
default 3994 (Ultra-2/2 CPUs/256 MB), no recommendations

 This is the systemwide number of processes available. You should leave sufficient space to the parameter maxuprc. The value of this parameter is influenced by the setting of maxusers.

 

maxuprc
default -5 (here: 3989), no recommendations

 This parameter describes the number of processes available to a single user. The actual value is determined from max_nprocs which is itself determined by maxusers. The negative value seems to be a relative distance with regards to max_nprocs, but I haven't been able to test this (yet).

 

npty
default 48, no recommendations

 The parameter defines the maximum number of BSD ttys (/dev/ptty??) available. A few BSD networking things might need these devices. If you run into a limit, you may want to increase the number of available ttys, but usually the size is sufficient.

 

pt_cnt
default 48, min 48, max 3000, no recommendations

 Solaris only allocated 48 SYSV pseudo tty devices (slave devices in /dev/pts/*). On a server with many remote login, or many open xterm windows you may reach this limit. It is of little interest to webservers or proxies, but of greater interest for personal workstations.

 

vac_size
default 16384 (with maxusers 249), recommended: don't set

 This parameter specifies the size of the virtual address cache. If a personal workstation with many open xterms and sufficient tty devices has a very degraded performance, this parameter might be too small. My recommendation is to let the system chose the correct value. The current value is determined by the size of maxusers.

 

ufs_ninode
default 4323 = 17*maxusers+90 (with maxusers 249), min 226, max 34906,

no recommendations

 The current parameter specifies the size of the inode table. This is some kind of cache. The actual value will be determined by the value of maxusers.

 Some webcache users increase this value. If your intention was to keep the inode for each squid data file in memory, forget it. You'd need over L1 * L2 * swapfiles_per_dir entries. But an increase with regards to the ncsize value might help a tiny little.

 

ncsize
default 4323 = 17*maxusers+90 (with maxusers 249), min 226, max 34906,

no recommendations

 This parameter specifies the size of the directory name lookup cache. A large directory name lookup cache size significantly helps NFS servers that have a lot of clients. On other systems the default is adequate.

 I don't know about the ties to ufs_ninode, but the formula is the same. The current value is determined by maxusers.

 I have heard from a few people who increase ncsize to 30000 when using the Squid webcache. Image, a Squid uses 16 toplevel directories and 256 second level directories. Thus you'd need over 4096 entries just for the directories. It looks as if webcaches and newsserver which store data in files generated from a hash need to increase this value for efficient access. Twice the default should be a good starting point. You may want to increase ufs_ninode by the same size, too.

 

ndquot
default 6484, no recommendations

 This parameter specifies the size of the quota table. Many standalone webservers or proxies don't use quotas.

 

nstrpush
default 9, no recommendations

 This parameter determines how many STREAMS modules you are allowed to push into the Solaris kernel - I guess this is a per user or per process count. The only application of widespread use which may need such a kernel module is xntp. Even with other modules pushed, usually you have sufficient room and no need to tweak this parameter.

 

strmsgsz
default 65536, no recommendations

 This parameter determines the maximum size of a message which is to be piped through the SYSV STREAMS.

 

strctlsz
default 1024, no recommendations

 The maximum size of the control part of a STREAMS message.

 

bufhwm
default 2 % of main memory, no immediate recommendations

 Now, considering the SVR3 buffer cache described by Maurice Bach, this parameter specifies the maximum memory size allowed for the buffer cache. The 0 value reported by sysinfo says to take 2 % of the main memory for buffer caches. sysdef -i shows the size in bytes taken for the buffer cache.

 I have seen Squid admins increasing this value up to 10 %. If you change this value, you have to enter the number of kByte you want for the buffer cache.

 

use_mxcc_prefetch
default 0 (sun4d) or 1 (sun4m), recommended: see text

 Adrian Cockroft explains in What are the tunable kernel parameters for Solaris 2? this parameter. The parameter determines the external cache controller prefetches. You have to know your workload. Applications with extensive floating point arithmetic will benefit from prefetches, thus the parameter is turned on on personal workstations. On random access databases with little or no need for float point arithmetic the prefetch will likely get into the way, therefore it is turned off on server machines. It looks as if it should be turned off on dedicated squid servers.

 

Some services use a multitude of caches files like Squid or some News server where names (URLs or articles) are mapped by a hash function to a shallow directory tree, helping the buffer cache and inode caches of the host file system (compared to using unlimited subdirectories like the CERN cache). As well-known in software engineering, the speedup by using the right algorithm usually far exceeds anything you can achieve by fiddling with the hardware or tweaking system parameters. Thus, a new storage scheme for mapped caches should provide food for thought.

 

7.3 100 Mbit ethernet related entries

Mr. Nebel and Mr. Hüsemann were so kind to give me a few hints concerning 100 Mbit ethernet interfaces and Solaris. It looks as if these cards default to halfduplex operations. In order to switch to full duplex mode, make sure your router can also work full duplex.

 

hme:hme_adv_100fdx_cap
default 0, recommended 1

 This parameter switches on the full duplex mode. Only use this parameter together with the next option.

 

hme:hme_adv_100hdx_cap
default 1, recommended 0

 Switch off the half duplex mode, must be used together with the previous parameter.

 

hme:hme_adv_autoneg_cap
default 1, recommended ?

 This parameter determines wether the SUN workstation should automatically negotiate the 100 Mbit with the switch or router. Usually Cisco switches also do auto negatiation, thus is may be necessary to set this switch to 0.

 

A few conditions on incorrectly working 100 Mbit interfaces result in a downgrade to 10 Mbit ethernet, or half-duplex mode. Thus check at all available ends, if you are really getting the data rate you are expecting.

 

7.4 How to find further entries

There are thousands of further items you can adjust. Every module which has a device in the /dev directory and a module file somewhere in the kernel tree underneath /kernel can be configured with the help of ndd. Wether you have to have superuser priveleges depends on the access mode of the device file.

 For instance, there exists a device /dev/hme and a kernel module /kernel/drv/hme. This driver is connected, as you might know, to the 100 Mbit ethernet interface. If you want to know what value you can tweak, you can ask ndd:

ndd /dev/hme \?
Of course, you can only change entries marked for read and write. If you tweaked enough and want to store some configuration as a default at boot time, you can enter your preferred values into the /etc/system file. Just prefix the key with the module name and separate both with a colon. You did see this earlier in the subsection on 100 Mbit ethernet and the System V IPC page.

 There is another way to get your hands on the names of keys to tweak. For instance, the System V IPC modules don't have a related device file. This implies that you cannot tweak things with the help of ndd. Nevertheless, you can obtain all clear text strings from the module file in the kernel.

strings -a /kernel/sys/shmsys # possible
nm /kernel/sys/shmsys # recommended
There is a number of strings you are seeing. Most of the strings are either names of function within the module or clear text string passages defined within. Strings starting with shminfo are the names of user tuneable parameters, though. Now, how do you separate tuneable parameters from the other stuff? I really don't know. If you have some knowledge about Sun DDI, you may be able to help me to find a recommendable way, e.g. using _info(9E) and mod_info.

 

8. Recommended patches

It is utterly necessary to patch you Solaris system, if you didn't already do so! Have a look at the DFN CERT patch mirror or the original source from SUN. There may be a mirror closer to you, e.g. EUNet and FUNET have their own mirrors, if I am informed correctly.

 In order to increase your TCP performance, security of websites and fix several severe bugs, do patch! Whoever still runs a Solaris below 2.5 should upgrade to 2.5.1 at least. I am about to find out how good Solaris 2.6 really is, and it is looking very promising.

 Please remember to press the Shift button on your netscape navigator while selecting a link. If the patch is not loadable, probably a new release appeared in the meantime. To determine the latter case, have a look at the directories of DFN CERT or SUN . The README file on the DNF-CERT server is kept without a version number and thus always up to date.

 

ip and ifconfig patch
103630-09 for Solaris 2.5.1 (README)

103169-12 for Solaris 2.5 (README)

 

tcp patch (only with ip patches)
103582-15 for Solaris 2.5.1 (README)

103447-09 for Solaris 2.5 (README)

 

Any system administrator should know the contents of SUN's patch page. Besides previously mentioned patches for a good TCP/IP performance, you should always consider the security related patches. Also, SUN recommends a set of further patches to complete the support for large IP addresses. You should really include any DNS related patch.

 The SUN supplied patches to fix multicast problems with 2.5.1 are incompatible with the TCP patch. Unfortunately, you have to decide between an unbroken multicast and a fixed TCP module. Yes, I am aware that multicast is only possible via UDP, nevertheless the multicast patch replaces the installed TCP module. If you have problems here, ask your SUN partner for a workaround - he will probably suggest upgrading to 2.6.

 

9. Related books and software

This section started after receiving some information from Christian Grimm and Franz Haberhauer on TCP/IP and performance related literature.

 

10. Uncovered material

There are a bunch of parameters which I didn't cover in the sections above, but some of which may be worth looking at:

 

tcp_ip_abort_linterval
default 180000

 linger?

 

tcp_ip_notify_cinterval
default 10000

 The timeout after which IP is notified by TCP to find a new route during an active open.

 

tcp_ip_notify_interval
default 10000

 The timeout after which IP is notified by TCP to find a new route for an established connection.

 

tcp_rexmit_interval_extra
Since 2.6: default 0

 Something in connection with retransmissions.
 

 

11. Startup scripts

For the important tweakable parameters exist startup scripts for Solaris. Only the first script is really necessary.
  1. The first script changed all parameters deemed necessary and described in the previous sections. The file should be called something like /etc/init.d/your-tune and you must link (hardlinks preferred, symbolic links are o.k.) /etc/rcS.d/S31your-tune to the init.d file.
  2.  Please read the script carefully before installing. It is a rather easy shell script. The piping and awking isn't as bad as it looks:

     

    Always tune the parameters to your needs, not mine. Thus, examine the values closely.

     

  3. The second script just changes the MTU of le0 from the IPX to the IEEE 802.3 size. The meaning is shown further up. The script is not strictly necessary, and reports about odd behaviour may have ceeded with a patched 2.5.1 or a 2.6.
  4.  If you intend to go ahead with this script, the file is called /etc/init.d/your-tune2 and you need to create a link to it (hard or soft, as above) as /etc/rc2.d/S90your-tune2. Please mind that GNU awk is used in the script, normal awk does not seem to work satisfactorily.