<div>I forgot to add. The version of SIPp used was <strong>1.1rc6</strong> (win32 version) from the Sourceforge website - which according to the site is the latest stable release. Most of the testing I did that time using SIPp was based on the uac scenario (
uac.xml - <font color="#000099"><span class="syntax13">Basic</span><span class="syntax13"> </span><span class="syntax13">Sipstone</span><span class="syntax13"> </span><span class="syntax13">UAC scenario)</span></font> as I have not played around with it extensively yet.
<br> </div>
<div>I think I may have to re-do some of these tests again with SIPp anyway to see if there are any other settings that might affect the results and maybe try and get some more formalised results.</div>
<div> </div>
<div> </div>
<div>Merry Christmas to All.</div>
<div><br> </div>
<div><span class="gmail_quote">On 12/17/06, <b class="gmail_sendername">Klaus Darilion</b> <<a href="mailto:klaus.mailinglists@pernau.at">klaus.mailinglists@pernau.at</a>> wrote:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">Hi MAx!<br><br>make sure to use newest sipp. AFAIR there were several patches to sipp CVS<br>version for high performance testing (for sure I know thre were TCP
<br>problem. Maybe there were also UDP problems too)<br><br>Further I would be interesting if the same problem happens when using TCP<br>too. In my tests TCP achieved higher performance then UDP (UDP's<br>retransmissions kills the proxy when getting to the performance limit -
<br>like an DoS attack).<br><br>regards<br>klaus<br><br>On Fri, December 15, 2006 23:10, Jiri Kuthan said:<br>> if you happen to have a PCAP file with the incident, let me please know.<br>><br>> -jiri<br>><br>
> p.s. even if you didn't tweak timers, the results may be suboptimal<br>> because<br>> the software version you are using is having rather indeterministic timer<br>> subsystem.<br>> For example, the recent measurements
<br>> (<a href="http://www.iptel.org/ser/doc/performance">http://www.iptel.org/ser/doc/performance</a>)<br>> show quite scattered server responsiveness under high load. (Note though<br>> that<br>> the measurement results were achieved in a best-effort manner based on the
<br>> tester<br>> knowledge and understanding of openser and that the result are not<br>> officially confirmed<br>> by the OpenSER project.) Whether it is indeed the cause is not certain<br>> though<br>> -- this looks really like a stealth bug.
<br>><br>><br>> At 15:58 15/12/2006, Max Gregorian wrote:<br>>>Thanks very much for all the replies. I shall try and post a config and<br>>> traces as soon as I can get them from the office.<br>>>
<br>>>Some more information, if it helps:<br>>><br>>>Server specs:<br>>>- HP ProLiant DL360 G4 (1U rack servers)<br>>>- 3 GHz processors (800 MHz FSB)<br>>>- 1 GB RAM<br>>>- 10K rpm SCSI HDs (in a RAID 1+0 Mirror)
<br>>><br>>># Servers are running OpenSER 1.0.1 (no-TLS).<br>>># Servers are listening on 3 ports (both tcp and udp for each port), so<br>>> in openserctl ps I am seeing 4 child processes for each port.
<br>>># Servers running CentOS Linux 4.3<br>>># MySQL installed when CentOS was installed but not running and not<br>>> currently being used with Openser.<br>>><br>>><br>>>Things I have pretty much managed to eliminate are:
<br>>>1. It's doesn't seem to be hardware. The specs for the servers are more<br>>> than sufficient I think.<br>>>2. It doesn't seem to be traffic/load related as I see these problems on<br>>> 2 brand new servers I have just installed with no traffic on them.
<br>>> However, it does seem to get worse with more traffic.<br>>>3. I don't think it's database related as I have deliberately not<br>>> configured mysql on any of the servers in case of database performance.
<br>>>4. I haven't played with the timers at all so far.<br>>>5. I haven't configured nscd yet, but as far as I can tell it's not<br>>> caching DNS.<br>>>6. Though openser is listening on tcp ports as well, currently only the
<br>>> udp ports are being used as most of our customers use hardware phones. In<br>>> any case, I haven't as yet seen as requests on tcp.<br>>>7. I am not sure it is DNS as in the tests I ran I sent requests directly
<br>>> to the external IP of the server and not to the domain name it is<br>>> responsible for. Also the test servers are now only responsible for one<br>>> domain, but in future will have more than one.
<br>>>8. Also TTL on the domain name is really short. Ping from the server<br>>> itself TTL=64 and ping times are low as you would expect (< 1ms when<br>>> pinging from the server itself). Ping from outside the network (from the
<br>>> internet - for me - tp the domain was) 12ms (average), no packet loss,<br>>> TTL = 53.<br>>>9. I have not setup any internal DNS entries for the domain. Servers are<br>>> resolving domain from entries in /etc/hosts.
<br>>><br>>>Like I said, it doesn't happen all the time - just maybe once or twice<br>>> every hour on the servers with more traffic.<br>>><br>>>I ran SIPp pointing at one of the new servers last week and at around
<br>>> 100CPS I was seeing about 2,000 out of approx. 10,000 calls were failing.<br>>> Setup was UAC -> openser -> UAS (Both UAC and UAS were running on the<br>>> same machine, but different ports). Again there is no traffic on these
<br>>> servers now so I have no idea why so many failed calls.<br>>><br>>>I am not sure if any of this information helps, but I am certainly open<br>>> to suggestions on things to try.<br>>><br>
>>Thanks in advance.<br>>><br>>><br>>><br>>>On 12/14/06, samuel <<mailto:<a href="mailto:samu60@gmail.com">samu60@gmail.com</a>><a href="mailto:samu60@gmail.com">samu60@gmail.com</a>
> wrote:<br>>>It might be due to a DNS query....whenver a request has to be<br>>>forwarded to a domain, openSER makes a DNS query to resolv the IP.<br>>>During this operation, the child processing the request will not
<br>>>answer to further incoming messages.<br>>><br>>>it also can be happening due to a spiral loop that stays on the server.<br>>><br>>>Without further information (confg,logs) it's hard to tell which is
<br>>>the reason...<br>>><br>>>hope it helps,<br>>>Samuel.<br>>><br>>>without more information<br>>><br>>>2006/12/14, Max Gregorian<br>>> <<mailto:<a href="mailto:gregorian442@googlemail.com">
gregorian442@googlemail.com</a>><a href="mailto:gregorian442@googlemail.com">gregorian442@googlemail.com</a>>:<br>>>> Hi all,<br>>>><br>>>> Just wondering if anyone else has had this problem. I have noticed
<br>>>> while<br>>>> tracing on my OpenSER server, that every now and then the server<br>>>> receives a<br>>>> packet which it does to respond to immediately, resulting in a string<br>>>> of
<br>>>> packets being sent to the server and then the server responding a few<br>>>> seconds later. This does not happen all the time, just say maybe once<br>>>> or<br>>>> twice every hour. The rest of the time the signaling is correct and
<br>>>> responses follow request packets in the correct order.<br>>>><br>>>> What I am trying to figure out is whether this is a load traffic issue<br>>>> (i.e.<br>>>> can the server not handle too much load), and if so is it OpenSER or
<br>>>> the<br>>>> network or the server in general? I have run diagnostics on the servers<br>>>> and<br>>>> there is nothing wrong with the hardware.<br>>>><br>>>> On the other hand Could this be related to any timer issues? I remember
<br>>>> there was mention of timers in SER but are there any default timer<br>>>> settings<br>>>> that can be tweaked?<br>>>><br>>>> Thanks in advance for any response.<br>>>>
<br>>>><br>>>> _______________________________________________<br>>>> Users mailing list<br>>>> <mailto:<a href="mailto:Users@openser.org">Users@openser.org</a>><a href="mailto:Users@openser.org">
Users@openser.org</a><br>>>> <a href="http://openser.org/cgi-bin/mailman/listinfo/users">http://openser.org/cgi-bin/mailman/listinfo/users</a><br>>>><br>>>><br>>>><br>>><br>>>
<br>>>_______________________________________________<br>>>Users mailing list<br>>><a href="mailto:Users@openser.org">Users@openser.org</a><br>>><a href="http://openser.org/cgi-bin/mailman/listinfo/users">
http://openser.org/cgi-bin/mailman/listinfo/users</a><br>><br>> --<br>> Jiri Kuthan <a href="http://iptel.org/~jiri/">http://iptel.org/~jiri/</a><br>><br>><br>> _______________________________________________
<br>> Users mailing list<br>> <a href="mailto:Users@openser.org">Users@openser.org</a><br>> <a href="http://openser.org/cgi-bin/mailman/listinfo/users">http://openser.org/cgi-bin/mailman/listinfo/users</a><br>>
<br><br><br></blockquote></div><br>