Random lockups on 5x series
Hi,
i have just deployed 30x 53i and 2x 57i + 560M sidecar at a brand new installation.
All phone are running latest released firmware from Aastra.
We are seeing random lockups on these phones and in particular the 57i. This typically happens after a page to all the phones is performed. Once a phone has locked up, only a hard reset will get it back again.
Has anyone else experienced this?
M
and it happens only with the 5xi + 560M combos, tried CTs and bland 57i too, the volume of calls is really high on those phones and they can crash from 5-6 times a day..
these messages are spammed in my CLI from these specific 5xi + 560M combos :
-- Incoming call: Got SIP response 489 "Bad Event" back from xxx.xxx.xxx.xxx
I opened a ticket at Aastra and the tech told me that the bug has probly rolled back in the 5xi's from a previous firmware version because it was earlier fixed on 480i series. Also gave me their Logging tool so I can send hem some stuff to work on.
I'll get back to you later if I get anything,
That's exactly the message I have been seeing as well.
I'll open a ticket with Aastra - I have to say I have been highly dissapointed with their new 5x line. Support has been good, but it looks like there are still a number of issues to be resolved with them and they are not as solid as their previous models.
Please kep us posted if you hear anything and I'll do likewise.
I am having the same issue on a 100 Phone install. The 9133 are fine but the 55i and 57i CT lock up. I get the same error message. I am also having BLF issues with the 57i CT with the 536 Side Card. The BLF's dont always show the right information. Lots of times they show a caller is on a call even though they have not been on a call for awhile and it shows people are not on a cal when they are on a call. Is anyone else having this problem.
I've taken this issue up with Aastraand they want me to capture phone log information. This is all very well, but not exactly something I want to do in a production environment.
I can typically cause one of the phones to lockup by doing pages one after the other in a short period of time. This generates a fair amount of SIP traffic, especially for the 57i and their BLF's.
My suspician is that the phones cannot handle the network traffic and end up losing or dropping SIP packets. Because of the lost packets, the states of various extensions are unknown.
This really shouldn't be happening on a released product, especially with so few connected sets.
Are you running any beta firmware on your phones?
M
I have a deploy of 80x53i and 2x57i+plus the advanced expansion module. The 57i have a heavy load of calls and at the moment I have experienced a couple of lockups during a period of two weeks. I have sent the logs from the phones to aastra support and told me they are working on it. The problem is that they are users migrating from old pbx systems to IP and they are not happy with the change. Hope aastra could release soon a fix for this problem.
regards.
I wish the Aastra guys would jump in here and help us out. Once again I am having BLF issues at another customers and my ticket has yet to be resolved with Aastra. The 55i will show some exts as being busy when my Snom's and Polycoms show that the ext is not busy and no one is using the ext that Aastra shows as being busy. I am going to have to start switching out my Aastras for another brand if they do not fix this issue soon.
I noticed that my 57i phone sends 489 Bad Event only When it gets a NOTIFY with an empty "Event" header such as "Event: -". Is this a known issue with Trixbox/Asterisk? I am running Trixbox 2.0.0/Asterisk 1.2.13.
Scenario is that NOTIFYs from Trixbox are coming correctly and BLFs are working fine as well But out of blue, Trixbox start sending NOTIFYs with "Event: -".
NOTIFY captured in my log looks like this:
NOTIFY sip:1263@100.100.100.101:5060;transport=udp SIP/2.0
Via: SIP/2.0/UDP 100.100.100.100:5060;branch=z9hG4bK3e480350;rport
From:
To: Alford
Contact:
Call-ID: 735cb46d9c95805a
CSeq: 120 NOTIFY
User-Agent: Asterisk PBX
Max-Forwards: 70
Event: -
Content-Type: unknown
Subscription-State: active
Content-Length: 0
This is a bug in Asterisk, but they don't seem to be interested in fixing it - see bug 7608
A workaround was put in for this in build 2.0.2 so the phone can recover from this problem. The phone will send "489 Bad Event" in response to the NOTIFY and will then send a new SUBSCRIBE message for the BLF key.
We have 12 57i's and are having the issue with BLF data being out-of-sync. If the phone loses communication with the server for even the briefest moment the BLF is hosed until you reboot the phone. Duplicate by restarting the asterisk service from the command line...
Has anyone found any kind of work-around for this? Is this a confirmed problem with the Aastra phones only? I love the Aastra script functionality, but this biggy behavior is a serious show stopper and we WILL change brands if necessary...
We have 12 57i's and are having the issue with BLF data being out-of-sync. If the phone loses communication with the server for even the briefest moment the BLF is hosed until you reboot the phone. Duplicate by restarting the asterisk service from the command line...
Has anyone found any kind of work-around for this? Is this a confirmed problem with the Aastra phones only? I love the Aastra script functionality, but this buggy behavior is a serious show stopper and we WILL change brands if necessary...
I am still trying to track this problem down, but do have slightly more information.
The sequence that I can determine goes like this -
First indication is that the receptionist starts having problems transferring calls.
Phone appears to respond slowly.
"sip show peers" response time for the reception extension increases.
this continues until the phone will not respond at all and only a hard restart will correct.
Sometimes only a restore to factory defaults will be needed to corrrect the problem (until the next time).
According to Aastra, they need the Stack Trace when this happens (which one can get via the web interface), however, when the phone locks up, one cannot get into the web interface. Kind of like a chicken and egg situation....
Needless to say, this is extremely frustrating for the receptionist and callers.
Please, if you are experiencing this issue - let Aastra know and ensure that a troubleticket is opened.
I emailed support with screen shots of the task/status page and sent a syslog of the failed dial events. I had a ticket open on this prior to 2.1.1 being released. I agree about the slowness of the phone when it is having problems. About 5 mins later the phone was working fine and dialing/transfering fine.
I have a complete box of this aastra crap phone ( 480i ) that have this hang problem with message 489 from SIP and I can't beleive that they released new phone's series with same problems , BLF problem with asterisk was also there 1 year ago. Look into Goole Groups Aastra-480i users and you will know what I'm talking about . For anyone that will claim that they have 1000's of their phone working correctly I say you fuck you because you're probably a employee of Aastra or someone paid to watch forum and post false setup.
Lets keep it civilized....
I too am frustrated, but Aastra is working on tracking the problem down. Having been in the IT business for quite a while, I know how difficult it can be to track these issues down. From my conversations with higher ups at Aastra, they are taking this problem very seriously and have a numbre of resources working on a fix.
In fact, I have it on good authority that a potential fix for the lockups is in the works and being tested right now.
Now whether this corrects some other issues mentioned in the thread (BLF, etc), I don't know.
Note: I do do not work for Aastra!
M
They told me the same thing about hang problem 1 year ago , and now you can see that they release a complete new line of phone with exactly the same problems , so if this was so much important to their eyes why their effort was put on new product line? Simply because they used the General Motor approach , put the same crap in a brand new metal box with good looking exterior and you will sell a new ton of it ( sunbird,sunfire,g5 you know?! ) , but anyway just the time will tell us how much they will be able to do that without losing customers but after 2 to 3 generation of the same buggy products you can be assure that market share will drop badly for Aastra. Interest by some 480i and 480i CT ?
They told me the same thing about hang problem 1 year ago , and now you can see that they release a complete new line of phone with exactly the same problems , so if this was so much important to their eyes why their effort was put on new product line? Simply because they used the General Motor approach , put the same crap in a brand new metal box with good looking exterior and you will sell a new ton of it ( sunbird,sunfire,g5 you know?! ) , but anyway just the time will tell us how much they will be able to do that without losing customers but after 2 to 3 generation of the same buggy products you can be assure that market share will drop badly for Aastra. Interest by some 480i and 480i CT ?
Whilst I agree to the fact that some problems have been hanging around for a while I'd rather work with Aastra to get the problem fixed for everyone. Everytime I have dealt with their support team I have been impressed.
The only criticism I have of Aastra in general would be that maybe they should put some more of the useful features into the firmware, like being able to use different ring tones instead of ring patterns or ring cadences, before they embarked on more leading edge stuff.
Having a pop at them gets us nowhere. What would be nice is if Aastra opened up their feature and bug lists so we could see if the issues we are having have been reported and any workarounds or ETA's on new features. Being open with the end user or support person, getting it in the neck from customers, would be a great move I think.
As for the 57i lockups, yes it's driving me nuts as I have several of these units and I get daily complaints about them not working. Moaning at Aastra doesn't get the problems fixed any faster as they need logs and other info in order to point the finger at the offending code.
If a few of us are having the same problem why doesn't Aastra get us all on a mailing list so we can discuss it and maybe one of us will hold the syslog to the fix!
Better communication and openess would certainly help in getting issues addressed and resolved in a more timely manner, but in my experience, this is normally not the case with "closed source" or proprietary products.
Coming from the open source world, I am used to joining mailing lists, pouring through bug reports and contributing where I can, in a collabortive manner. Companies, on the other hand, are more used to handling things internally and keeping things quiet, especially "negative" things, like bugs, complaints and problems.
I would love to see Aastra open up their own community forums, give us access to their knowledge base and bug reports, but I don't think that is going ot happen anytime soon.
I am impressed that at least they are aware of the community and some of their employees have signed up to the forum and post every once in a while - it is a good first step.
Getting back to the issue at hand - I've had some test firmware running for the last few days without an issue on one of the 57i's and I'm about to upgrade another one. No problems so far.
M
As far as I understand, this fixes a potential lockup issue with the 57i and sidecar.
I managed to capture a stack trace showing the suspended process, Aastra was eventually able to replicate the problem, which I believe was a memory leak, hence the new test firmware.
I have gone the pedantic=yes route a while back - I think it solved a Cseq out of order error but not much else.
As far as the "Bad Event" message, I believe I still received these prior to the firmware upgrade, but haven't seen it since, although this is not an indication that it is resolved.
M
I have had the Bad Event messages since 2.0.2 I seem to remember. It's good if they have seen a suspended process cos everytime I check the phone after a lockup the stack looks OK. The problem does seem to only be relevent to the 57i with a sidecar attached so here is hoping!!
Just to add to the woes. Yesterday and today I have had 3 57i's with 560 sidecars lockup and become fairly useless. They were all running the 2.1.1 firmware. I've tried to downgrade to 2.0.2 but I'm being told that they are still not working.
Has anyone else had this sort of problem recently?
I have two 57i phones with sidecars at receptionist desks. Latest firmware from last time i checked this week. I have about 20 480i phones at employee/executive desks. THe 480i phones work flawelessly. I really do mean flawlessly. THe 57i phones are rebooted probably once every few hours. My problems?
1) They suddenly dont ring at all during incoming calls.
2) Transferred calls drop for no reason during the xfer.
3) Intercom randomly stops working (using intercom feature through freebx, server side in phone)
rebooting helps most of the time.
I'm considering replacing them with Cisco SIP phones or Polycom or something with a proven track record of being expensive and stable :)
So bad news.... there is still an issue with the latest test firmware.
We have not had anymore lockups (which is good), however, one of the 57i's started displaying "No Service" for no reason. Only a reboot of the phone corrected it.
Checking the logs, we see the
Got SIP response 489 "Bad Event"
and then after that, lots of
Got SIP response 500 "CSeq Number Out of order"
I still have the
pedantic = yes in my sip.conf which was supposed ot solve the Cseq errors.
Previously, when the phones would lockup, these messages were NOT generated.
So, I believe there is more than 1 issue with these phones and sidecar.
1. Random lockups - phone starts acting slowly, eventually complete lockup. This is indicated by suspended tasks in the stack trace (if you can get there in time to see them). Potentially fixed in latest test firmware.
2. 489 "Bad Event" leading to disconnectin from the Asterisk server. Unsure what causes this. Still to be investigated further. No solution at present.
Don't know where to go from here..... just thought I'de share more info...
M
Yes same problems here. I have a site with 50 x 53i's and 1 x 57 with side car, and we are pretty much having all the issues you guys are. Lockups, NO SERVICE occaisionally (on the 53is too)
57i is getting bad event message. Also getting reports of calls disconnecting, and problems with transfer of some calls from 57i with sidecar.
I have a smaller site with about 20 handsets, (mainly 480i's) and 2 x 57i's with side cars, and have have no problems reported from that site yet. It has been installed for about a month now.
We are setting up a reboot script to reboot at midnite, to see if this helps. As best we can tell, we are only getting a few (problem transfers) a day, and as far as we have been told, the phone has only had "no service" lockup twice in a month or two since we installed it.
The side car is comming off soon, and if worse comes to worse, we will put in a 480i as switchboard.
The occaisional "no Service" on the 53i's is a worry too.
We are running 2.1.1 on all phones.
I have multiple site that have the 57i w/ sidecar lockups 3 at 1 site, 2 at another and 3 out of 4 at another. All are on the same firmware (latest).
Is there any response from Aastra about what to do here?
-Mike Donato
mdonato@sitetechllc.com
I would think after reading the last 4 pages, they have plenty of tickets. We are sending them one anyway, but I got a sneaky suspicion they are well aware of this problem.
What would be nice is a bit of feed back in here from Aastra about where they are at with this, and some possible patches or work arounds would be nice too.
This is exactly what we are experiencing as writtin by mag. What else do you need to know.
So bad news.... there is still an issue with the latest test firmware.
We have not had anymore lockups (which is good), however, one of the 57i's started displaying "No Service" for no reason. Only a reboot of the phone corrected it.
Checking the logs, we see the
Got SIP response 489 "Bad Event"
and then after that, lots of
Got SIP response 500 "CSeq Number Out of order"
I still have the
pedantic = yes in my sip.conf which was supposed ot solve the Cseq errors.
Previously, when the phones would lockup, these messages were NOT generated.
So, I believe there is more than 1 issue with these phones and sidecar.
1. Random lockups - phone starts acting slowly, eventually complete lockup. This is indicated by suspended tasks in the stack trace (if you can get there in time to see them). Potentially fixed in latest test firmware.
2. 489 "Bad Event" leading to disconnectin from the Asterisk server. Unsure what causes this. Still to be investigated further. No solution at present.
Mustardman... I do have multiple open tickets.
As far as why multiple sidecar phones....why not is my response. In smaller offices, there isn't always a set receptionist. Sometimes there's 2-3 people who answer phones and cover for each other. (for example: an admin assistant, a sales assistant and maybe a book keeper). At other offices, I have people who want them for simply speed dials and other functions. All in all though, why would/should it matter if there are more than 1 sidecar in an office?
I'm just posting here to see if there is any update to the never ending "we're working on it" response. I really love these phones and I want to feel confident when I put them out in the field.
-Mike Donato
mdonato@sitetechllc.com
Yeah I realize that, what I was trying to confirm, was on the same sites that the 5 series phones are failing, are the 480is failing too? I am desperate to get a work around for my reception phone, to the ppoint I am prepared to drop in a 480i with no sidecar, so that we dont get lost calls.
I heard a whisper today that aastra may be releasing another firmware version in the next few weeks anyway. If this does not work I am probably going to try a linksys 962, with 932 sidecar/s which I beleive now works BLF with asterisk.
the call interrupt thing on the aastra phone stiull bugs me anyway. Their fudgey solution of putting in an "ignore" key as a work around is still a bit crappy IMO anyway.
I have a busy receptionist, complaining about this interrupt to dialing anyway.
I need to do some testing, but I think the call interrupt thing effects the operator when trying to transfer a call. If this is the case then this phone is just not suitable as a console anyway.
Setting the incomming call interrupts dialing to off just presents an igonre key, it still interupts the dialing.
Again, what are your log files telling you???????
What part of "failing ONLY with sidecars" did you not understand. You continually state that 5xi's are failing and wondering if the 480i's are failing too implying it's a standalone phone issue. If you just want to get a standalone phone why not.......remove the sidecar!!!!!!
I'm getting a sneaky suspicion some of you are just trying to throw gas on the flames. I'd like to trace some of your IP addresses and see where they go.
Just an unhappy VAR that is rapidly switching to Polycoms because of the lack of attention. It is clear that there is a serious reliability problem. I have 1 ticket open that's been open for 5 months. Haven't heard word one, and there is no reason to open another. It's the same issue.
I've had 2 57i's now go bad in 5 months since the install. (Repaired under warranty, but nonetheless) .
Mind you, I use 57's at home. Love the phones. Just really nervous at this point rolling them to clients.
-Darren
Mustard man. Chill out!
I think you are completley misunderstanding me. Listen to me again.
I am still in a learning phase on deeper IP stuff, but I have been working with PABX's for 30 years, so I know how phones should behave.
I can tell you for 100% certain fact it is not just a 57i issue. We have 50 x 53is on the same site also losing calls, in termittantly, and a few that are going into "no service" intermittantly. No real rhme or reason to this. we have triple checked the phone configs, and set factory defaults too.
They also send the same error messages but not as much. I have a very good engineer who has checked logs and knows exactly what is happening. We just dont know why. It's a firmware issue.
This is not just a 57i issue. It presents itself more on the 57i with sidecar because of the sidecar. Our main problem has been calls dropping out. Not lots of them, but enough to dent the confidence of our client in the product. They are spooked, and now everything is becomming an issue. Interestingly we set a midnite reboot script on the main console about 2 days ago, and have no reported call drops on the 57i since then. A bit early to say wether that has helped, but is encouraging.
We see the same sip errors from 53i's but it is not as much of a problem, only because the phones are not in use as much. It is a known memory leak problem I am lead to believe on very good authority.
Like I said earlier in another post, Santa may be deliveing some new firmware real soon I also hear.
What my original point is again, does anyone have a site with 30 or more phones in the 5 series, and perhaps a few 480is' just to see if they are suffering the same problems. I KNOW the 480i does not have a sidecar, what I am looking for is something that might not drop calls or lockup. Bugger the side car. Looking for a suitable workaround at this point
My other point was to do with normal behavior of the phone, if normal is the right word, aside from any lockups. You get a busy receptionist, who has incomming calls crashing into attempted transfers, you get a pretty frustrated operator.
Dont get me wrong I really like aastra phones, apart from a few annoyances like this.
Becuase of the interruption to outbound or transferring calls, I may not use them as a console again. I am going to try an spa962 with a sidecar, because I know the call is not interupted whilst dialing.
This is not such a problem for the average user, but it IS a problem for a busy receptionist.
These are still my preffered phones but horses for courses.
I've been running some test firmware for the last few weeks and have not experienced any lockups on hte 57i, although "No Service" still appears from time to time.
I've also heard a new testing release is coming real soon.
At the moment, I am recommending the Snom 360 + Sidecar for all installations involving a receptionist or busy phone. They do not interrupt the current call when another call comes in and they work well with the parking lot (don't have to hangup the phone once parked). Only thing is, I patched Asterisk with the metermaid patch to provide hints for parked calls.
I don't have any installations with significant numbers of both 5xi phones and older 480i models, so I cannot tell whether the issues are experienced with both phones. I do have a number of sites using 480i's and they have not experienced the no service issues as the 5xi's.
M
Snoms are very stable, But, the supervised transfer method is also horrible like :
Press hold
press BLF key to call transferring party
press transfer to release call.
Aastra is better at least:
press transfer
press blf key, and hangup to release call.
That's nearly what it should be, apart from annoying call interrupt when another call comes in and confuses the hell otu of them.
Ideal world.
Press blf key
Talk
hangup to release call, or go back to active line.
IP land just do not get transfers I am afraid. I am looking at linksys with 932. Blind transfer is great, assuming that is the norm, which it isnt. But supervised transfer is:
Press transfer
Press BLF
Press xfer line to release, or go baclk to caller byt pressing active line key.
I actually found that using a softkey with asterisk transfer code on the aastra is pretty good way of transferring. Just need to code it as a speed dial.
Call stays on the same line which is mus less confusing for an operator not used to having transferring call go to a second line.
BUT interupt call still confuses the issue, as you can be in th process of transferring a call, and you have to press ignore key to proceed. At least using speed dial key with asterisk means the original call is on the line it cam in on, so one less flashing line for the operator to deal with.
I would like to know if the linksys, can have a key programmed that does the same thing by sending dtmf "in call" that might be a soilution I think.
just saw there is a new firmware out. amazing, I didn't even get an update to my ticket to notify. How poor is that.
-Mike Donato
mdonato@sitetechllc.com
The new 2.1.2 firmware should fix all the expansion module related lockups that people have been experiencing. For the record they were all related these being busier phones, hence more likely running into the various issues mentioned in the release notes - ie it will benefit everyone using a 5i phone to upgrade, not just people with expansion modules.
If people still experience problems after upgrading, please report them to Aastra support - this is the only way we have of gathering the required information to fix the problems (which we are committed to doing).
I hope to release a new RPM on trixbox with this firmware early in the New Year.
Great that you have finally released this update and that it SHOULD fix the issues we have been experiencing, but I still contend that there is a better way of going about this than currently.
For now, the process appears to be that a person experiences an issue, they report it and go through the Aastra process of getting debug information, getting more information, and more, and then finally proving it is an Aastra issue, all without any input or confirmation from anyone else. Then the next person comes along with the same issue, faces the usual denials from Aastra, goes through the above process and maybe eventually gets an acknowledgement that there is an issue and they will work on it. And then the next person, and the next, etc.
How about opening up your bug tracking database to people so that we can communually solve problems and help both ourselves and Aastra? This is one way that open source and the community really works.
I am not asking Aastra to open source their code and let anyone and everyone have access, all I am asking is that you give us some credit and take the issues we face seriously and allow us to help you resolve any issues we come across. We like your phones, we want to use and sell more of them, but don't underestimate us and our experiences. We (most of the time) know what we are talking about!
Take a look at the Snom site and see how they provide information about their products (not to say that they couldn't improve either). Be more open and share, it will only benefit Aastra in the end.
M
I second the comment about opening up the bug tracker. I've had many issues and email conversations with Aastra about the 57i problems and even though the support I have received has been very good it always feels like you are alone. Seeing that other people are having the same problems is good!
I have to admit that I have not come across any more issues since upgrading, but then my issues were random and only occurred once every week or two.
I am not totally convinced that the issues have been resolved until I can run at least 2 months without a lockup.
M
Just thought of something - have you confirmed that you have no network issues at all?
I've found the Aastra's to be particularly susceptable to problems on the network and have come across this at 2 different clients.
Easiest way to check is to do a netstat -su
make sure you have no Udp packet receive errors.
M
Almost a week for me an no phone lockups yet on my control site. They only have a skeleton staff on at the moment, so I have not other transfer problems reported as yet. They will be back to full staff next week so I will get a better idea by the end of next week. Least we dont seem to be getting any "no service" phones after about a week,
OK.... I've got a syslog of the problems but the customers site is 2.30hrs drive away so I can't just pop over to investigate. I'll go through the syslog and I'm hoping I can get it solved as I've been told the manager had a bit of a paddy and 'ripped' the phone out.
I had a report of a phone going into no service yesterday about a week after the upgrade. Have not heard the results of the syslog yet. I will let you know here when I find out what happened. It does seem to be much better so far though. Before we were getting anything up to 10 phones out of 50 per day going into no service. This usually happened in extended periods of inactivity like overnite or weekends. This one happened at 3 o'clock in the arvo at the 57i reception phone which has'nt happened sso far. Could be something else I hope

Member Since:
2006-05-31