I can try and re-run, how many would you recommend worth trying for this scenario ?
Thanks Seb Sebastian Wagner Director Arrakeen Solutions, OM-Hosting.com http://arrakeen-solutions.co.nz/ https://om-hosting.com - Cloud & Server Hosting for HTML5 Video-Conferencing OpenMeetings <https://www.youracclaim.com/badges/da4e8828-743d-4968-af6f-49033f10d60a/public_url> <https://www.youracclaim.com/badges/b7e709c6-aa87-4b02-9faf-099038475e36/public_url> On Tue, 2 Feb 2021 at 21:56, Maxim Solodovnik <[email protected]> wrote: > Have you tried to increase maxThreads for Tomcat? > > On Tue, 2 Feb 2021 at 15:26, [email protected] <[email protected]> > wrote: > > > I doubled it to 4GB OpenMeetings and 4GB KMS. I updated the docker > instance > > to run Openmeetings with xms=2GB and Xmx=4GB. > > > > And I did run exactly the same test again: > > - 50-60 users > > - staggered to enter in a time period around 5-10min > > - distributed into 10 conference rooms 4x4 and 2 webinars with 20 users > > each > > - each test runs calls the API to login/createRoomHash and then load the > > URL with the room (plus start webcam/audio stream in the conference > rooms) > > > > The results look almost the same. There is hardly any improvement: > > > > - CPU still spikes to almost 100%, memory is not a problem > > - Empty video pods as well as video pods where webcam stream didn't > > start > > > > There isn't a crash, but that is mostly because I stagger it to enter the > > server over a 5-10min period. Which didn't crash the 2GB instance either. > > > > Comparison of the CPU graphs of both hardware configuration and test > runs: > > > > > https://cwiki.apache.org/confluence/display/OPENMEETINGS/Performance+Testing#PerformanceTesting-ClusterPerformancetestresult02-022021 > > > > There is pretty much no improvement. > > > > There is some work on the application side needed. This does not look > like > > getting better by throwing more hardware at it. > > > > It is really quite limiting to have no logs about any sort of performance > > indicators like call length to narrow down where the bottleneck is. > > You may find some very low hanging fruits in terms of optimisation if you > > can simply concentrate on the top ten calls and optimise those. > > Rather than looking at CPU and memory graphs. > > > > Thanks > > Sebastian > > > > Sebastian Wagner > > Director Arrakeen Solutions, OM-Hosting.com > > http://arrakeen-solutions.co.nz/ > > https://om-hosting.com - Cloud & Server Hosting for HTML5 > > Video-Conferencing OpenMeetings > > < > > > https://www.youracclaim.com/badges/da4e8828-743d-4968-af6f-49033f10d60a/public_url > > > > > < > > > https://www.youracclaim.com/badges/b7e709c6-aa87-4b02-9faf-099038475e36/public_url > > > > > > > > > On Tue, 2 Feb 2021 at 17:18, [email protected] < > [email protected]> > > wrote: > > > > > Have we ever looked into which java method would require the most > > > resources/time during the process of entering the conference room ? > > > > > > Sebastian Wagner > > > Director Arrakeen Solutions, OM-Hosting.com > > > http://arrakeen-solutions.co.nz/ > > > https://om-hosting.com - Cloud & Server Hosting for HTML5 > > > Video-Conferencing OpenMeetings > > > > > > < > > > https://www.youracclaim.com/badges/da4e8828-743d-4968-af6f-49033f10d60a/public_url > > > > > > < > > > https://www.youracclaim.com/badges/b7e709c6-aa87-4b02-9faf-099038475e36/public_url > > > > > > > > > > > > On Tue, 2 Feb 2021 at 16:48, Maxim Solodovnik <[email protected]> > > > wrote: > > > > > >> While do load testing I did the following: > > >> > > >> create Jmeter test loading "semistatic" stateless error page with 300 > > >> simultaneous threads (I can share this test it is very simple) > > >> CPU usage of OM process was near to 100% > > >> the situation is better if Tomcat has more threads (maxThread > parameter) > > >> > > >> I guess we need to check "The Ultimate Tomcat Performace Guide" :))) > > >> > > >> On Tue, 2 Feb 2021 at 10:41, [email protected] < > > [email protected] > > >> > > > >> wrote: > > >> > > >> > Also the spikes are on the CPU actually more than on the memory: > > >> > > > >> > > > >> > > > https://cwiki.apache.org/confluence/display/OPENMEETINGS/Performance+Testing#PerformanceTesting-ClusterPerformancetestresult02-022021 > > >> > > > >> > The spike is just 50-60 users. > > >> > > > >> > Why would CPU spike to almost 100% just for that amount of users ? > > >> > > > >> > I can try with 4GB for Openmeetings and repeat the test. > > >> > > > >> > Thanks > > >> > Seb > > >> > > > >> > Sebastian Wagner > > >> > Director Arrakeen Solutions, OM-Hosting.com > > >> > http://arrakeen-solutions.co.nz/ > > >> > https://om-hosting.com - Cloud & Server Hosting for HTML5 > > >> > Video-Conferencing OpenMeetings > > >> > < > > >> > > > >> > > > https://www.youracclaim.com/badges/da4e8828-743d-4968-af6f-49033f10d60a/public_url > > >> > > > > >> > < > > >> > > > >> > > > https://www.youracclaim.com/badges/b7e709c6-aa87-4b02-9faf-099038475e36/public_url > > >> > > > > >> > > > >> > > > >> > On Tue, 2 Feb 2021 at 16:34, Maxim Solodovnik <[email protected] > > > > >> > wrote: > > >> > > > >> > > On Tue, 2 Feb 2021 at 10:30, [email protected] < > > >> > [email protected]> > > >> > > wrote: > > >> > > > > >> > > > I think what you mean is you have OpenMeetings and MySQL and KMS > > on > > >> one > > >> > > > instance with 4GB. > > >> > > > > > >> > > > But its 2GB Just for OpenMeetings. > > >> > > > > > >> > > > > >> > > I mean > > >> > > 4GB just for OM (demo-next) > > >> > > 8GB just for OM (demo-prod) > > >> > > and this might need to be increased in case of many users > > >> > > > > >> > > Additionally Tomcat's maxThreads might need to be increased here: > > >> > > > > >> > > > > >> > > > >> > > > https://github.com/apache/openmeetings/blob/master/openmeetings-server/src/main/assembly/conf/server.xml#L74 > > >> > > > > >> > > I suspect lot's of simultaneous users need more resources > > >> > > > > >> > > > > >> > > KMS is separated with another 2GB > > >> > > > MySQL is on another server with another 2GB > > >> > > > So that would be 6GB in total. But only 2 are allocated to > > >> > OpenMeetings. > > >> > > > > > >> > > > XmX=2GB for OpenMeetings should be enough and not crash with > 50-60 > > >> > users > > >> > > > entering the room at the same time. > > >> > > > > > >> > > > Thanks > > >> > > > Sebastian > > >> > > > > > >> > > > Sebastian Wagner > > >> > > > Director Arrakeen Solutions, OM-Hosting.com > > >> > > > http://arrakeen-solutions.co.nz/ > > >> > > > https://om-hosting.com - Cloud & Server Hosting for HTML5 > > >> > > > Video-Conferencing OpenMeetings > > >> > > > < > > >> > > > > > >> > > > > >> > > > >> > > > https://www.youracclaim.com/badges/da4e8828-743d-4968-af6f-49033f10d60a/public_url > > >> > > > > > > >> > > > < > > >> > > > > > >> > > > > >> > > > >> > > > https://www.youracclaim.com/badges/b7e709c6-aa87-4b02-9faf-099038475e36/public_url > > >> > > > > > > >> > > > > > >> > > > > > >> > > > On Tue, 2 Feb 2021 at 16:26, Maxim Solodovnik < > > [email protected] > > >> > > > >> > > > wrote: > > >> > > > > > >> > > > > Hello Sebastian, > > >> > > > > > > >> > > > > It seems 2GB of RAM is not enough for OM > > >> > > > > `OutOfMemoryError: Container killed due to memory usage` > > >> > > > > I never use less than 4GB (8-16GB in production) > > >> > > > > > > >> > > > > > > >> > > > > > > >> > > > > On Tue, 2 Feb 2021 at 09:54, Maxim Solodovnik < > > >> [email protected]> > > >> > > > > wrote: > > >> > > > > > > >> > > > > > > > >> > > > > > > > >> > > > > > On Tue, 2 Feb 2021 at 07:23, [email protected] < > > >> > > > > [email protected]> > > >> > > > > > wrote: > > >> > > > > > > > >> > > > > >> Hi, > > >> > > > > >> > > >> > > > > >> I have been conducting a few more performance and load > tests > > >> with > > >> > > the > > >> > > > > goal > > >> > > > > >> of increasing participants to 100++. > > >> > > > > >> > > >> > > > > >> The challenge is: > > >> > > > > >> *If more then 50-60 users dynamically create a room Hash > > (using > > >> > > > > Soap/Rest > > >> > > > > >> API) and use that Hash to enter the conference room CPU and > > >> memory > > >> > > > > spikes > > >> > > > > >> and server crashes* > > >> > > > > >> > > >> > > > > > > > >> > > > > > Can you share API call sequence? > > >> > > > > > Maybe we can write JMeter scenario for this? > > >> > > > > > > > >> > > > > > server crash is something bad > > >> > > > > > What is happening? Is it a JVM crash? Or is the system low > of > > >> > > resources > > >> > > > > > and the kernel kills the trouble-maker? > > >> > > > > > > > >> > > > > > > > >> > > > > >> *Test scenario observations:* > > >> > > > > >> - It does not matter if those users try to enter the same > > >> room or > > >> > > > > >> separate > > >> > > > > >> rooms. In the above test scenario it's a mix of 4x4 > > conference > > >> > rooms > > >> > > > and > > >> > > > > >> 20x1 webinars > > >> > > > > >> - This can be reproduced stable and repetitively > > >> > > > > >> - The issue starts with API calls taking 10sec++ and > getting > > >> more > > >> > > > > slower. > > >> > > > > >> Until the OpenMeetings Tomcat instance crashes > > >> > > > > >> - The issue also manifests that -BEFORE- the server > crashes > > >> you > > >> > can > > >> > > > see > > >> > > > > >> video pods not completing the initialisation in the > > conference > > >> > room > > >> > > > > >> itself. > > >> > > > > >> For example missing video pods or video pods without a > webcam > > >> > > stream. > > >> > > > > >> Likely to be linked to slow running API or web-socket calls > > >> > > > > >> => I can deliver data samples or screenshots if required > via > > >> our > > >> > > > > >> confluence > > >> > > > > >> space. > > >> > > > > >> > > >> > > > > >> *Hardware and software:* > > >> > > > > >> - Server and OpenMeetings Instance is isolated on a > > separated > > >> > > > hardware > > >> > > > > >> and > > >> > > > > >> has 2GB of memory allocated > > >> > > > > >> - There is no spike on KMS or Database > hardware/CPU/memory. > > >> The > > >> > > spike > > >> > > > > is > > >> > > > > >> only in the OpenMeetings Tomcat Server instance > > >> > > > > >> > > >> > > > > >> *Possible ways to mitigate without code changes:* > > >> > > > > >> - You can mitigate part of this issue if you spread the > > users > > >> to > > >> > > > enter > > >> > > > > >> over a longer time period. However it needs more than 10min > > >> > > separation > > >> > > > > to > > >> > > > > >> enter without issues for 50-60 participants > > >> > > > > >> - You can mitigate part of this issue if you for example > > >> create > > >> > the > > >> > > > > >> room-hash in a different process (like 1h before using) and > > >> once > > >> > all > > >> > > > > >> hashes > > >> > > > > >> are created you enter the conference room. It still leads > to > > >> > issues, > > >> > > > but > > >> > > > > >> you can enter up to 100 users within 5-10min, if you just > use > > >> the > > >> > > > links, > > >> > > > > >> rather than create the link AND entering with the link at > the > > >> same > > >> > > > > >> time/process > > >> > > > > >> - Increasing Tomcat to more than 2GB of memory per Tomcat > > >> > instance > > >> > > > may > > >> > > > > >> help, not sure by how much though > > >> > > > > >> > > >> > > > > >> => I think we should spend further time and propose ways > to > > >> get > > >> > rid > > >> > > > of > > >> > > > > >> those spikes. The mitigations are not realistic to really > be > > >> able > > >> > to > > >> > > > use > > >> > > > > >> in > > >> > > > > >> practise. > > >> > > > > >> > > >> > > > > >> *My proposal is:* > > >> > > > > >> There is further analysis needed: > > >> > > > > >> - Capture all OpenMeetings calls that happen during the > > create > > >> > room > > >> > > > > hash > > >> > > > > >> and conference room-enter > > >> > > > > >> - Measure call lengths and any calls during the create > room > > >> hash > > >> > > and > > >> > > > > >> conference room-enter and specific CPU spikes or memory > usage > > >> > based > > >> > > > on a > > >> > > > > >> per call basis > > >> > > > > >> - Eventually get a stack trace or have a profile available > > >> that > > >> > > > exports > > >> > > > > >> the current in memory objects to review where and what > create > > >> > those > > >> > > > > spikes > > >> > > > > >> > > >> > > > > >> Once a per-call analysis is there it should be a lot more > > easy > > >> to > > >> > > > > pinpoint > > >> > > > > >> specific issues and propose improvements. > > >> > > > > >> > > >> > > > > >> As with all performance optimisation this is likely to need > > >> more > > >> > > > > >> discussion > > >> > > > > >> once more detailed data is available. > > >> > > > > >> > > >> > > > > >> Thanks, > > >> > > > > >> Sebastian > > >> > > > > >> > > >> > > > > >> Sebastian Wagner > > >> > > > > >> Director Arrakeen Solutions, OM-Hosting.com > > >> > > > > >> http://arrakeen-solutions.co.nz/ > > >> > > > > >> https://om-hosting.com - Cloud & Server Hosting for HTML5 > > >> > > > > >> Video-Conferencing OpenMeetings > > >> > > > > >> < > > >> > > > > >> > > >> > > > > > > >> > > > > > >> > > > > >> > > > >> > > > https://www.youracclaim.com/badges/da4e8828-743d-4968-af6f-49033f10d60a/public_url > > >> > > > > >> > > > >> > > > > >> < > > >> > > > > >> > > >> > > > > > > >> > > > > > >> > > > > >> > > > >> > > > https://www.youracclaim.com/badges/b7e709c6-aa87-4b02-9faf-099038475e36/public_url > > >> > > > > >> > > > >> > > > > >> > > >> > > > > > > > >> > > > > > > > >> > > > > > -- > > >> > > > > > Best regards, > > >> > > > > > Maxim > > >> > > > > > > > >> > > > > > > >> > > > > > > >> > > > > -- > > >> > > > > Best regards, > > >> > > > > Maxim > > >> > > > > > > >> > > > > > >> > > > > >> > > > > >> > > -- > > >> > > Best regards, > > >> > > Maxim > > >> > > > > >> > > > >> > > >> > > >> -- > > >> Best regards, > > >> Maxim > > >> > > > > > > > > -- > Best regards, > Maxim >
