| |
| News & Reviews |
Welcome to the Voxilla VoIP Forum.
Voxilla has been a trusted source for accurate, up-to-date information on the IP Communications industry since 2002. A dedicated staff of reporters and engineers produce feature articles and product reviews to keep industry watchers abreast of the people, companies, and trends driving a fast moving market.
You are currently viewing our boards as a guest which gives you limited access to view most discussions and access our other features. By joining our free community you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is fast, simple and absolutely free so please, join our community today!
If you have any problems with the registration process or your account login, please contact contact us.
Voxilla VoIP Forum |
AAH 2.5 Memory leak? reboot = 39% - Day 2 = 75% - helpTechnical support, how-to guides, troubleshooting, and general assistance, from beginner to seasoned pro, this is where to discuss Asterisk, the most powerful open source PBX. |
| | LinkBack | Thread Tools | Rate Thread | Display Modes |
| |||
| I am looking for advice on how to track down what is causing physical memory usage to increase. This is causing dialparties.agi application to take longer to run, which results in the incoming call answer being delayed longer and longer. I am running an AMD 380 with 384 mem. When I first boot it shows physical mem at 39% - and all runs fine. Come the next day the physical mem is up to 65% and the problem starts. |
| |||
| I am having the same problem, but my server has 2gigs so I can go longer before it starts to dog down the server. But hey, If I wanted to reboot every 3rd day, I would have run this on windows (j/k I love Windows 2003 as much as Linux)... still a problem - hopefully if others have it they will post up.
__________________ [iSEPIC] |
| |||
| I was wondering if this would peak at some mem level. Seeing that your running 2G and still encountering the problem has squashed that hope. I tried running top to see what processes are running but from first looks that didn't provide me any more insight. I was thinking that I need something that keeps a running log of when applications start, and what their resource utilization profile looks like each hour. That would help me to know where to start looking. Also, I don't use Sugar CRM so maybe I could pair the application down to bare asterisk. Your thoughts.... |
| |||
| I'm not using sugar crm, or anything really more than default install, except for fax, and tif to pdf convert... I did however, change my sendmail config files to use a proxy - with the 2 gigs, after a fresh reboot it uses about 135MB, after being up about 3 days now,its up to 558MB of physical ram usage... I do use the AAH web page quite a bit, along with rebooting asterisk (amportal stop, amportal restart) - when I do the amportal stop/start - it comes back using about 1MB more ram than seconds ago - odd that it doesn't clear out all the ram. Perhaps it's something else. I was thinking about setting up, as you said, something to capture usage.. I found this page to be helpful http://mail.nl.linux.org/linux-mm/2003-03/msg00077.html - so maybe even just running the PS command as he stated >> into a log file every 10 mins or so over the few days and see what grows.
__________________ [iSEPIC] |
| |||
| |
| | |
| |||
| I was looking into whether I have introduced a bug by using the stealth attendant code as described at nerd vittles. This dialplan has incoming calls forwarded to an extension that isn't defined using amportal. You put an entry in the from-internal context to manage calls directed to that extension. I have modified my code somewhat from what he suggests but have been over that code time and time again. I don't have the fax support that you mentioned, so other than the auto attendant, and some other abilities to accept external sip calls, it is a basic install... Do you think capturing PS will do anything? From memory I don't think it gives you mem usage -- or am I using the wrong cmd line ps -ax |
| |||
| I managed to get a bit different behaviour - Physical mem was up to 70% and the answer delay kicked in but I did manage to get the answer delay back to the desired state by doing one or more of the following: 1 - Reset MySql (no change to answer delay) 2 - Re-read configs (made a change in the hang-up macro to not add a 5 sec delay when hanging up a fake extension - read up on the nerd vittles stealth attendant routine if you want to know more about fake ext) With this, the answer delay disappeared.. I did notice today though that when an incoming call was answered, the person sounded very choppy when speaking at low volume. I think there are a few things eating up CPU cycles and suspect that my physical mem usage is probably at 95%+. |
| |||
| I've been looking more and more at mine too.. seems when I play a 20meg mp3 file as my hold music (and I play music on hold for all incoming calls to both my phone lines (one for me, one for my roommate) ----- so I'm also using the native music on hold player with asterisk at home 2.5 (1.2.4) - if I don't use the MOH, no memory leak (well not nearly as bad anyways). If I do, it shoots up quickly. My next steps will to try the non native ( mpg123 or whatever it's called).. if that doesn't work, I'll go back to the crappy streaming music for now or try smaller mp3 files).
__________________ [iSEPIC] |
| |||
| I too have multiple MOH and answer greetings depending on caller ID. I probably have 12 different songs in 3 different MOH categories. I don't recall the file sizes but other than the default songs, each MP3 is encoded down to 8khz and play for approx 30 - 60 seconds. Seems like the mem increase hasn't been as dramatic over the last two days. Maybe something I did with fixing an error in my extensions_custom.conf file (skipped a priority number). Will keep an eye on it. As per the choppy incoming audio reported earlier, it appears to be limited to incoming calls via my PSTN trunk and not via the VOIP trunk. I can't see a relationship that points back to the Asterisk server so I'm thinking this is an isolated issue, rooted in my sipura SPA3000 fxo adapter. |
| | |
| Thread Tools | |
| Display Modes | Rate This Thread |
| |
| | ||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Need to reboot SPA-2100 few times a day | DownUnder | Linksys (Sipura) VoIP Support Forum | 1 | March 23rd, 2006 07:46 PM |
| AAH 2.5 VMWARE Physical Memory = 100% ???? | ravi28UK | Asterisk Support Forum | 5 | February 24th, 2006 08:55 PM |
| Memory Leak?? | cagab | Asterisk Support Forum | 2 | November 30th, 2005 09:56 PM |
| Linksys WRT54G reboot causes AAH to lose IAX2 registration | bbbeavis | Asterisk Support Forum | 2 | August 30th, 2005 05:41 AM |
| I've had enough - BV number down all day | deacom | BroadVoice Support Forum | 1 | July 29th, 2004 11:15 PM |