Kosala, OK, so this says that the subscriber you're trying to register as can't be found.
The stress nodes will by default use DNs from 201000000 onwards. What numbers did you provision? The docs (https://github.com/Metaswitch/crest/blob/dev/docs/Bulk-Provisioning%20Numbers.md) recommend provisioning 2010000000-2010099999. However, I recall you trying to run "sudo ./BulkProvision homer 21000 29999 example.com 7kkzTyGW", which would only provision the numbers 21000-29999. Maybe there's a mismatch here? Matt -- Matt Williams Lead Architect, Project Clearwater +44 (0) 20 8366 1177 From: Kosala [mailto:[email protected]] Sent: 17 August 2016 14:34 To: Matt Williams (projectclearwater.org) <[email protected]> Subject: Re: [Project Clearwater] Clearwater Digest, Vol 40, Issue 24 I am getting followings in sip-stress_18438_errors.log: 2016-08-17 12:26:26:258 1471436786.258684: Aborting call on unexpected message for Call-Id '[email protected]<mailto:[email protected]>': while expecting '401' (index 3), received 'SIP/2.0 404 Not Found^M Via: SIP/2.0/TCP 10.0.22.193:5060;rport=8826;received=10.0.22.193;branch=z9hG4bK-18438-898-2-2010003204-^M Record-Route: <sip:scscf.cwaio:5054;transport=TCP;lr;service=scscf;billing-role=charge-orig>^M Call-ID: 2010003204///[email protected]<mailto:[email protected]>^M From: <sip:[email protected]<mailto:sip%[email protected]>>;tag=18438SIPpTag00898^M To: <sip:[email protected]<mailto:sip%[email protected]>>;tag=z9hG4bKPjc1xOWCH0ooae6O9npl7fSSrQ-x0vmoV9^M CSeq: 1 REGISTER^M Content-Length: 0^M Any advice is appreciated. Many Thanks, Kosala On Wed, Aug 17, 2016 at 2:20 PM, Kosala <[email protected]<mailto:[email protected]>> wrote: Hi Matt, I can see this scenario screen from sip-stress.1.out. I hope this will be the stress test output results. ------------------------------ Scenario Screen -------- [1-9]: Change Screen -- Users (length) Port Total-time Total-calls Remote-host 2500 (0 ms) 5060 77.15 s 2500 10.0.21.192:5060(TCP) Call limit reached (-m 2500), 1.002 s period 1 ms scheduler resolution 2175 calls (limit 2500) Peak was 2499 calls, after 0 s 0 Running, 2177 Paused, 8 Woken up 0 dead call msg (discarded) 0 out-of-call msg (discarded) 3 open sockets Messages Retrans Timeout Unexpected-Msg Pause [0ms/10:00] 2500 0 REGISTER ----------> 326 0 401 <---------- 0 0 0 325 REGISTER ----------> 0 0 200 <---------- 0 0 0 0 REGISTER ----------> 0 0 401 <---------- 0 0 0 0 REGISTER ----------> 0 0 200 <---------- 0 0 0 0 Pause [ 4:50] 0 0 REGISTER ----------> B-RTD1 0 0 200 <---------- E-RTD1 0 0 0 0 REGISTER ----------> B-RTD1 0 0 200 <---------- E-RTD1 0 0 0 0 Pause [ 4:50] 0 0 Pause [$pre_call_delay] 0 0 INVITE ----------> B-RTD2 0 0 100 <---------- 0 0 0 0 INVITE <---------- 0 0 0 0 100 <---------- 0 0 0 0 INVITE <---------- 0 0 0 0 100 ----------> 0 0 180 ----------> 0 0 180 <---------- 0 0 0 0 Pause [ 6000ms] 0 0 200 ----------> 0 0 200 <---------- 0 0 0 0 ACK ----------> 0 0 ACK <---------- 0 0 0 0 UPDATE ----------> 0 0 UPDATE <---------- 0 0 0 0 200 ----------> 0 0 200 <---------- E-RTD2 0 0 0 0 Pause [ 24.0s] 0 0 BYE ----------> B-RTD3 0 0 BYE <---------- 0 0 0 0 200 ----------> 0 0 200 <---------- E-RTD3 0 0 0 0 Pause [$post_call_delay] 0 0 ------- Waiting for active calls to end. Press [q] again to force exit. ------- Last Error: Aborting call on unexpected message for Call-Id '1079-13729@... However there is an unexpected message error at the end. Thanks Kosala On Wed, Aug 17, 2016 at 12:40 PM, Matt Williams (projectclearwater.org<http://projectclearwater.org>) <[email protected]<mailto:[email protected]>> wrote: Kosala, They should log to /var/log/clearwater-sip-stress - in particular, the *.1.out file is a log of the output from sipp. Do you see anything there? Matt From: Kosala [mailto:[email protected]<mailto:[email protected]>] Sent: 17 August 2016 12:33 To: Matt Williams (projectclearwater.org<http://projectclearwater.org>) <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]> Subject: Re: [Project Clearwater] Clearwater Digest, Vol 40, Issue 24 Hi Matt, I followed above steps and run sudo apt-get install clearwater-sip-stress. How do I know stress tests are running? Is is said, stress tests start automatically but I cannot see it. Thanks Kosala On Tue, Aug 16, 2016 at 2:01 PM, Matt Williams (projectclearwater.org<http://projectclearwater.org>) <[email protected]<mailto:[email protected]>> wrote: Kosala, I'm glad you've made progress on this. I had a look at the crest package version, but it seems to be later than the most recent all-in-one image uses (1.0-160811.103635). When did you download the all-in-one image, and did you apply any patches to it? Regarding making calls for stress testing, this is where http://clearwater.readthedocs.io/en/latest/Clearwater_stress_testing.html#manual-i-e-non-chef-stress-runs comes in - you need to • create a separate VM • set the bono_hostname and count parameters in /etc/clearwater/shared_config • install the clearwater-sip-stress package. I hope that helps. BTW, please remember to copy in [email protected]<mailto:[email protected]> - that means that other people can see and comment on your posts. Thanks, Matt From: Kosala [mailto:[email protected]<mailto:[email protected]>] Sent: 16 August 2016 10:48 To: Matt Williams (projectclearwater.org<http://projectclearwater.org>) <[email protected]<mailto:[email protected]>> Subject: Re: [Project Clearwater] Clearwater Digest, Vol 40, Issue 24 Hi Matt, Bulk provisioning for all in one image is done using https://github.com/Metaswitch/crest/blob/dev/docs/Bulk-Provisioning%20Numbers.md. I have records in tables in homestead_cache and homer. Now How to make calls for stress testing? Is there any guideline? Many Thanks, Kosala On Tue, Aug 16, 2016 at 9:04 AM, Kosala <[email protected]<mailto:[email protected]>> wrote: Hi Matt, TBH, The document in https://github.com/Metaswitch/crest/blob/dev/docs/Bulk-Provisioning%20Numbers.md is not clear to me. Step 1. Homestead node in deployment. I do not know where is that node in all in one image. It will be helpful the location of that folder from the root. Step 2. A simple example will be helpful. Thanks Kosala On Tue, Aug 16, 2016 at 8:50 AM, Kosala <[email protected]<mailto:[email protected]>> wrote: Hi Matt, I downloaded the all in one package from http://clearwater.readthedocs.io/en/stable/All_in_one_OVF_Installation.html and imported using oracle virtual box on ubuntu 14.04, Crest version [cw-aio]ubuntu@cwaio:~$ sudo dpkg-query -s crest Package: crest Status: install ok installed Priority: optional Section: network Installed-Size: 40098 Maintainer: Project Clearwater Maintainers <[email protected]<mailto:[email protected]>> Architecture: amd64 Version: 1.0-160815.145007 Thanks Kosala On Mon, Aug 15, 2016 at 5:56 PM, Matt Williams (projectclearwater.org<http://projectclearwater.org>) <[email protected]<mailto:[email protected]>> wrote: Kosala, That's odd - I don't see that failure on my all-in-one image (spun up just now from the latest image) - which version of all-in-one-image are you running, and which version of crest does it include ("sudo dpkg-query -s crest")? You may be able to work-around this either by taking the latest all-in-one image or by following the simpler bulk-provisioning process at https://github.com/Metaswitch/crest/blob/dev/docs/Bulk-Provisioning%20Numbers.md (when this doc talks about using the sstables_provisioning process for very large sets of numbers, we mean 1M+ numbers - the 9000 you're provisioning are well within the capabilities of this process). I hope that helps - please let me know how you get on. Cheers, Matt From: Clearwater [mailto:[email protected]<mailto:[email protected]>] On Behalf Of Kosala Sent: 15 August 2016 16:58 To: [email protected]<mailto:[email protected]> Subject: Re: [Project Clearwater] Clearwater Digest, Vol 40, Issue 24 HI Matt, Thanks. I am trying to follow https://github.com/Metaswitch/crest/blob/dev/src/metaswitch/crest/tools/sstable_provisioning/README.md for bulk provisioning. When I am trying to execute sudo ./BulkProvision homer 21000 29999 example.com<http://example.com> 7kkzTyGW in all-in-one image, I am getting an error java.lang.UnsupportedClassVersionError: ClearwaterBulkProvisioner : Unsupported major.minor version 52.0 Java version info in all-in-one image: [cw-aio]ubuntu@cwaio:/usr/share/clearwater/crest/tools/sstable_provisioning$ java -version java version "1.7.0_101" OpenJDK Runtime Environment (IcedTea 2.6.6) (7u101-2.6.6-0ubuntu0.14.04.1) OpenJDK 64-Bit Server VM (build 24.95-b01, mixed mode) Thanks Kosala On Mon, Aug 15, 2016 at 2:17 PM, <[email protected]<mailto:[email protected]>> wrote: Send Clearwater mailing list submissions to [email protected]<mailto:[email protected]> To subscribe or unsubscribe via the World Wide Web, visit http://lists.projectclearwater.org/mailman/listinfo/clearwater_lists.projectclearwater.org or, via email, send a message with subject or body 'help' to [email protected]<mailto:[email protected]> You can reach the person managing the list at [email protected]<mailto:[email protected]> When replying, please edit your Subject line so it is more specific than "Re: Contents of Clearwater digest..." Today's Topics: 1. Stress Testing (Kosala) 2. Re: Stress Testing (Matt Williams (projectclearwater.org<http://projectclearwater.org>)) 3. Re: Stress Testing (Matt Williams (projectclearwater.org<http://projectclearwater.org>)) ---------------------------------------------------------------------- Message: 1 Date: Mon, 15 Aug 2016 10:38:57 +0100 From: Kosala <[email protected]<mailto:[email protected]>> To: [email protected]<mailto:[email protected]> Subject: [Project Clearwater] Stress Testing Message-ID: <CAGEe17an2B_bGj8NfKFwms+nU=vjuryht5ce+mcy1pjcm4y...@mail.gmail.com<mailto:vjuryht5ce%2bmcy1pjcm4yf%[email protected]>> Content-Type: text/plain; charset="utf-8" Hi Team, I installed clearwater "all-in-one" image on Oracle Virtual Box. 1. Can I do stress testing ? 2. How to do that? I am trying to follow Stress Testing on http://clearwater.readthedocs.io/en/stable/Clearwater_stress_testing.html#configuring-udp-stress . Using chef - > It seems chef/knife install only supports in the Cloud approach (The automated install is only supported for deployments running in Amazon?s EC2 cloud). Manual approach, It looks not supporting all in one image. Many Thanks Kosala -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.projectclearwater.org/pipermail/clearwater_lists.projectclearwater.org/attachments/20160815/15fc3b07/attachment-0001.html> ------------------------------ Message: 2 Date: Mon, 15 Aug 2016 09:54:02 +0000 From: "Matt Williams (projectclearwater.org<http://projectclearwater.org>)" <[email protected]<mailto:[email protected]>> To: Kosala <[email protected]<mailto:[email protected]>> Cc: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Subject: Re: [Project Clearwater] Stress Testing Message-ID: <cy1pr0201mb15636f1f20eb91bb180f053390...@cy1pr0201mb1563.namprd02.prod.outlook.com<mailto:cy1pr0201mb15636f1f20eb91bb180f053390...@cy1pr0201mb1563.namprd02.prod.outlook.com>> Content-Type: text/plain; charset="utf-8" Kosala, While it's theoretically possible to do stress testing against an all-in-one image, I'd recommend against it - the all-in-one image has been squeezed down a lot to fit on a single VM and we'd expect very low performance out of it, and I could believe that overload behavior might be suboptimal. If you're looking to stress test for performance reasons, I'd strongly recommend running a full deployment as this will give you much more reliable numbers. If you're looking to stress test for other reasons, please can you share a bit more about what you're trying to achieve, and we can have a think about what would work best? If you do really want to run stress testing against an all-in-one image, it should probably work, even though http://clearwater.readthedocs.io/en/latest/Clearwater_stress_testing.html#manual-i-e-non-chef-stress-runs only talks about Chef and manual installs. You'd need to set the "bono_servers" option in /etc/clearwater/shared_config to point to your all-in-one node, and I suspect you'd need to reduce the "count" option to something a lot lower - I'd start around 5000 and see how that goes. I hope that helps. Please let me know how you get on. Thanks, Matt From: Clearwater [mailto:[email protected]<mailto:[email protected]>] On Behalf Of Kosala Sent: 15 August 2016 10:39 To: [email protected]<mailto:[email protected]> Subject: [Project Clearwater] Stress Testing Hi Team, I installed clearwater "all-in-one" image on Oracle Virtual Box. 1. Can I do stress testing ? 2. How to do that? I am trying to follow Stress Testing on http://clearwater.readthedocs.io/en/stable/Clearwater_stress_testing.html#configuring-udp-stress. Using chef - > It seems chef/knife install only supports in the Cloud approach (The automated install is only supported for deployments running in Amazon?s EC2 cloud). Manual approach, It looks not supporting all in one image. Many Thanks Kosala -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.projectclearwater.org/pipermail/clearwater_lists.projectclearwater.org/attachments/20160815/29d6328d/attachment-0001.html> ------------------------------ Message: 3 Date: Mon, 15 Aug 2016 13:16:47 +0000 From: "Matt Williams (projectclearwater.org<http://projectclearwater.org>)" <[email protected]<mailto:[email protected]>> To: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Subject: Re: [Project Clearwater] Stress Testing Message-ID: <cy1pr0201mb156372246dc5a77b11839f9d90...@cy1pr0201mb1563.namprd02.prod.outlook.com<mailto:cy1pr0201mb156372246dc5a77b11839f9d90...@cy1pr0201mb1563.namprd02.prod.outlook.com>> Content-Type: text/plain; charset="utf-8" Kosala, You'll need to follow all the steps in http://clearwater.readthedocs.io/en/latest/Clearwater_stress_testing.html#manual-i-e-non-chef-stress-runs, i.e. bulk provisioning subscribers and installing clearwater-sip-stress, but that should be it. Just to be clear, though, you'll need to set the configuration and install the clearwater-sip-stress package on a separate VM, not on the all-in-one image itself - the all-in-one image almost certainly opens up ports that conflict with those that clearwater-sip-stress tries to use. This isn't a very well-trodden path, though, so you may hit further issues - let me know if so, and we can try to work through them. Cheers, Matt From: Kosala [mailto:[email protected]<mailto:[email protected]>] Sent: 15 August 2016 13:41 To: Matt Williams (projectclearwater.org<http://projectclearwater.org>) <[email protected]<mailto:[email protected]>> Subject: Re: [Project Clearwater] Stress Testing Hi Matt, Thanks. I set the bono_servers option. Should I add count =5000 to /etc/clearwater/shared_config ? Do I need other changes to run stress tests on all-in-one image? Thanks Kosala On Mon, Aug 15, 2016 at 10:54 AM, Matt Williams (projectclearwater.org<http://projectclearwater.org><http://projectclearwater.org>) <[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>> wrote: Kosala, While it's theoretically possible to do stress testing against an all-in-one image, I'd recommend against it - the all-in-one image has been squeezed down a lot to fit on a single VM and we'd expect very low performance out of it, and I could believe that overload behavior might be suboptimal. If you're looking to stress test for performance reasons, I'd strongly recommend running a full deployment as this will give you much more reliable numbers. If you're looking to stress test for other reasons, please can you share a bit more about what you're trying to achieve, and we can have a think about what would work best? If you do really want to run stress testing against an all-in-one image, it should probably work, even though http://clearwater.readthedocs.io/en/latest/Clearwater_stress_testing.html#manual-i-e-non-chef-stress-runs only talks about Chef and manual installs. You'd need to set the "bono_servers" option in /etc/clearwater/shared_config to point to your all-in-one node, and I suspect you'd need to reduce the "count" option to something a lot lower - I'd start around 5000 and see how that goes. I hope that helps. Please let me know how you get on. Thanks, Matt From: Clearwater [mailto:[email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>>] On Behalf Of Kosala Sent: 15 August 2016 10:39 To: [email protected]<mailto:[email protected]><mailto:[email protected]<mailto:[email protected]>> Subject: [Project Clearwater] Stress Testing Hi Team, I installed clearwater "all-in-one" image on Oracle Virtual Box. 1. Can I do stress testing ? 2. How to do that? I am trying to follow Stress Testing on http://clearwater.readthedocs.io/en/stable/Clearwater_stress_testing.html#configuring-udp-stress. Using chef - > It seems chef/knife install only supports in the Cloud approach (The automated install is only supported for deployments running in Amazon?s EC2 cloud). Manual approach, It looks not supporting all in one image. Many Thanks Kosala -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.projectclearwater.org/pipermail/clearwater_lists.projectclearwater.org/attachments/20160815/11781755/attachment.html> ------------------------------ Subject: Digest Footer _______________________________________________ Clearwater mailing list [email protected]<mailto:[email protected]> http://lists.projectclearwater.org/mailman/listinfo/clearwater_lists.projectclearwater.org ------------------------------ End of Clearwater Digest, Vol 40, Issue 24 ******************************************
_______________________________________________ Clearwater mailing list [email protected] http://lists.projectclearwater.org/mailman/listinfo/clearwater_lists.projectclearwater.org
