(Asterisk 10.9.0) How to fix errors in the logs after updating Askozia from version 4 to version 5?

Question

Asterisk

Good day!
After updating the program (installing a new version on a separate VM under Hyper_v and transferring the configuration), a number of problems appeared:
1) The main problem that I could not solve on my own was caused by the periodic disappearance of phone registration for 10 seconds. If you call the phone at this moment, a message is displayed that the phone is busy. If you make a call from this phone when the registration is lost, the call starts only after the remaining time until the new registration (1-10 seconds).
Error log:
2017-02-27 15:36:12 asterisk[4054]: NOTICE[23271]: chan_sip.c:21839 in handle_response_peerpoke: Peer '242' is now Lagged. (2171ms / 2000ms)
2017-02-27 15:36:12 asterisk[4054]: NOTICE[23271]: chan_sip.c:21839 in handle_response_peerpoke: Peer '224' is now Lagged. (2193ms / 2000ms)
2017-02-27 15:36:22 asterisk[4054]: NOTICE[23271]: chan_sip.c:21839 in handle_response_peerpoke: Peer '242' is now Reachable. (8ms / 2000ms)
2017-02-27 5:36:22 asterisk[4054]: NOTICE[23271]: chan_sip.c:21839 in handle_response_peerpoke: Peer '224' is now Reachable. (84ms / 2000ms) 2
) "Method: BYE" does not work after the call ends voice queue length queuing to Local/[email protected];2
2017-02-27 15:41:49 asterisk[4054]: WARNING[31746]: channel.c:1520 in __ast_queue_frame: Exceptionally long voice queue length queuing to Local/[email protected];2
2017-02-27 15 :42:27 asterisk[4054]: WARNING[32538]: channel.c:1520 in __ast_queue_frame: Exceptionally long voice queue length queuing to Local/[email protected];2
2017-02-27 15:42:28 asterisk[ 4054]: WARNING[32538]: channel.c:1520 in __ast_queue_frame: Exceptionally long voice queue length queuing to Local/[email protected];2
2017-02-27 5:42:31 asterisk[4054]: WARNING[23271 ]: chan_sip.c:3952 in __sip_autodestruct: Autodestruct on dialog '[email protected]:5060' with owner SIP/0501-0000424a in place (Method: BYE). Rescheduling destruction for 10000
3) Strange errors with the database. Perhaps because of this problem, call statistics at remote branches are not recorded. (I mean displaying information about missed/made calls on phone screens)
Error log:
2017-02-27 15:40:41 asterisk[4054]: ERROR[23243]: cel_sqlite3_custom.c:281 in write_cel: database is locked. MIKO_SQL: INSERT INTO cel (eventtype,eventtime,cidname,cidnum,cidani,cidrdnis,ciddnid,context,exten,channame,appname,appdata,amaflags,accountcode
And if I run an old PBX, then there are no such problems. All Hyper-V settings and even more so ats - identical.
(There are 2 virtual machines on one server, versions 4 and 5 of the PBX. They receive IP via DHCP - IP for the PBX is reserved in it. In order to avoid errors and not have to rewrite the new IP in all phones, I just cut down the old PBX, then prescribe on the DHCP poppy of the new PBX and at startup the correct IP address is picked up.I think I did not make any mistakes here).
In the global settings, Localnet is registered for internal phones, ports 5060 and the RTP range are forwarded on the firewall (you can see how the branches connect to the PBX, the voice goes both ways, this is also all right). And in general the old automatic telephone exchange works without problems.
You ask - "Why not work on the old PBX?" - Of course, I will transfer all the work of IP telephony to the old version without problems, but only if I do not find an answer within a week. There is also a reason that the authorities paid for the transition to the new version in order to use the new goodies, but that's another story ..
I ask you comrades for help, at least a hint where to dig, because the developer and seller of this program refused to help.
Sent them tcpdump - got the answer - "We don't know the cause of the problem"
Paid support answered the same.
As I tried to diagnose myself (I also sent all the results to the support):
1) I was looking for differences with the previous version, There are none, everything is identical and, according to the developers, when transferring the configuration, everything should work as before. Alas, this did not happen
2) I sent tcpdump from the PBX, they said everything was fine .. (I don’t know if this is bad or not, but the log ran into 20 megabytes of file.pcap in 15 seconds)
3) I didn’t receive an answer from ifconfig:
~ # ifconfig
eth0 Link encap:Ethernet HWaddr XX.XX.XX.XX.XX.XX
inet addr:192.168.0.78 Bcast:192.168.0.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:33243185 errors :0 dropped:760822 overruns:0 frame:0
TX packets:26377056 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:2502911363 (2.3 GiB) TX bytes:2809031324 (2.6 GiB)
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:65536 Metric:1
RX packets :7331066 errors:0 dropped:0 overruns:0 frame:0
TX packets:7331066 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:1471873209 (1.3 GiB) TX bytes:1471873209 (1.3 GiB)
4) The history of calls at remote branches has ceased to be recorded (xs is possibly evidence)
5) I re-read all the forums, including foreign ones. Eliminated all possible problems with DUDOS attacks, etc.
6) CPU and memory usage during rush hour no more than 10%. All conversations are recorded, everything is in order here too.
7) Phones and the PBX server work through a new gigabit switch (set to eliminate problems with the network)
PS tonight I will register the parameter srvlookup = no yes" I want to at least solve the problem with the registration dump. I'm also trying to delve into this topic www.ibm.com/developerworks/ru/library/l-hisock but in general to delve into the brains of the PBX according to the developer "It is highly not recommended" supposedly everything should work "out of the box" Global settings:
allowoverlap=no
bindport=5060
bindaddr=0.0.0.0
srvlookup=yes
useragent=AskoziaPBX
relaxdtmf=yes
alwaysauthreject=yes
videosupport=yes
notifybusy=yes
multicast_provisioning_interface=192.168.0.78
counteronpeer=yes
notifyhold=yes
notifycid=ignore-context
notifyringicing=yes
pedant yes
callcounter=yes
register_retry_403=yes
defaultexpiry=300
mineexpiry=60
maxexpiry=3600
snomautoprovisioningurl= http://192.168.0.78:56080/prov.php?mac={mac}
yealinkautoprovisioningurl= 192.168.0.78:56080
gigasetautoprovisioningurl=192.168.0.78:56080/gigaset
externip=XX.XX.XX.XX
localnet=192.168.0.0/24
; manual attributes
localnet=192.168.0.0/255.255.255.0 Localnet
is already set by default, I'll try to remove it from manual attributes
Perhaps some parameter was transferred incorrectly. Driving everything from scratch is the worst hemorrhoid, but if I don’t get to the bottom of the truth, I’ll have to ..
Let me remind you that the version of Asterisk 10.9.0 is for versions 4 and 5 of the program. And Askozia is just a graphical web interface for ease of use.
If you need any additional information, please indicate in your answer. Thanks in advance for your help!

Answer 1

1) The main problem that I could not solve on my own is caused by the periodic disappearance of phone registration for 10 seconds ...
Peer '242' is now Lagged. (2171ms / 2000ms)

Or remove qualify in "no" or look for plugs/differences in the network, tk. say that everything works on the "old" automatic telephone exchange.
Either there are differences, all the same, or somewhere something is buggy.
It is better to take out the rest of the questions separately, because you have written too much in one topic and it is inconvenient to answer.

Answer 2

Transferred everything manually, all problems are gone except for
one 00002bbb;2
2017-02-27 15:41:49 asterisk[4054]: WARNING[31746]: channel.c:1520 in __ast_queue_frame: Exceptionally long voice queue length queuing to Local/[email protected];2
2017-02 -27 15:42:27 asterisk[4054]: WARNING[32538]: channel.c:1520 in __ast_queue_frame: Exceptionally long voice queue length queuing to Local/[email protected];2
2017-02-27 15:42: 28 asterisk[4054]: WARNING[32538]: channel.c:1520 in __ast_queue_frame: Exceptionally long voice queue length queuing to Local/[email protected];2
2017-02-27 5:42:31 asterisk[4054]: WARNING[23271]: chan_sip.c:3952 in __sip_autodestruct: Autodestruct on dialog '[email protected]:5060' with place0 owner/040 Method: BYE). Rescheduling destruction for 10000
For a long time I tried to figure out what this is connected with, but I did not find an answer even on the off forum of the program. Providers tell them packets come, nothing is blocked, everything is ok with tcpdump too. Naturally, the supplier's support could not help with anything.
By independent tests, I found out that the problem occurs only when the call goes to the route (the Askozia program has a route module). And when the provider leaves immediately on the phone, everything is OK.
I think this is a bug in the routes module on versions 5.1 and 5.2.
Rolled back to 4.0.3I will wait for the next updates