SBC randomly changing online status

[litus]

Hi there,

at one of our customers 3CX Enterprise (hosted on the European 3CX servers) installation we have trouble with the local SBC.

The problem is, that the SBC status randomly (and then for something between a few minutes and a few hours) changes from up to down and back several times a minute. While this happens nobody behind the SBC can make a call (outbound or inbound).
In addition, sometimes it occurs that users that are working directly through the webclient will see the error that the connection has been lost (which usually would indicate network issues on the server side) but regarding 3CX there are no connection issues on their side.

What we’ve tried:
While this happens, we’ve
- restarted the local SBC (this didn’t help)
- restarted the router and the modem that are connecting the SBC to the internet (did not make any difference)
- Monitor the Local Internet connection to other servers (all fine)
- restart the service at the cloud installation (not always helping)
- gave full firewall permissions and port forwarding (no difference)
What we are currently trying:
- we monitor the hosted installations Ip to see if we can identify any downtime

***
out of the 3cxSystemService.log

------------|Inf(00)| Date: 3/27/2021 4:28:25 AM
2021/03/27 04:28:25.822|1332|0001|Inf|Log is continued
2021/03/27 04:28:25.823|1332|0001|Inf|Log is created
2021/03/27 04:28:27.688|1332|0001|Info|Log verbosity is set to Low

This seemed to be the primary incident:
04/27/2021 4:50:48 PM - SBC CONN EVENT: id=2, name=sbc.c0924627.Zrq4Qb2go1v8, client=Zrq4Qb2go1v8, ip=::ffff:217.###.###.#3 : Authenticated
04/27/2021 4:50:46 PM - Failure response: CSTA Recv 503/INFO from 127.0.0.1:5080 tid=4b9ad53952c99464 Call-ID=MkNiHlaem5xym_p6A-SvMg..:

SIP/2.0 503 Remote end of tunnel is not connected
Via: SIP/2.0/UDP 127.0.0.1:5060;branch=z9hG4bK-524287-1---4b9ad53952c99464;rport=5060
To: <sip:024@############:5060>;tag=2079718839
From: <sip:024@############.my3cx.de>;tag=f709bf7a
Call-ID: MkNiHlaem5xym_p6A-SvMg..
CSeq: 431 INFO
Warning: 399 #####-#####-#####-#####-#####.localdomain "Remote end of the bridge is not connected"
Content-Length: 0

/* Comment by the author
* The snipped below is repeated multiple times with differenet users which have been hide trough ####
*/
2021/03/30 05:00:32.394|1332|0057|Erro|Error while Office365SubscriptionManager.UpdateSubscription Create subscription for resource: Users/#######@#######.onmicrosoft.com/contacts changeTypeExpected: created,updated,deleted Exception: Code: UnknownError
Inner error:
AdditionalData:
date: 2021-03-30T03:00:32
request-id: 81ede3b6-f93d-4361-ab81-cce8e99d3120
client-request-id: 81ede3b6-f93d-4361-ab81-cce8e99d3120
ClientRequestId: 81ede3b6-f93d-4361-ab81-cce8e99d3120

2021/03/30 05:00:32.548|1332|0052|Erro|Error while running Office365DirectoryReader.GetContactsAsync for user #######@#######.onmicrosoft.com: Code: MailboxNotEnabledForRESTAPI
Message: The mailbox is either inactive, soft-deleted, or is hosted on-premise and the caller didn’t specify EHAM shard behavior.
ClientRequestId: 03c52b38-4628-46f7-ab87-5e5cccec428a
***

The before mention error happens for all users at the same time and is not restricted to certain users.
If anyone has some ideas or hints, this would greatly be appreciated!
Best,
Marius

 

[NickD_3CX]

Hi!

Just to make sure, you mean that the SBC is connected to a 3CX PBX that is "Hosted by 3CX" right? Could you please send me the FQDN so I can verify and see where the instance is located?

Also, one thing you could do is log into http://customer.3cx.com and go to the Instance Manager and check if there are an services down when this happens, of high CPU usage, or low disk space.

 

[litus]

I will check the instance manager next time the issue comes up.

We tried with the firewall by opening all ports but that did not help but we don't have access to the router. However, the ISP set it up as requested and I think there's no additional firewall interfering (LANCOM Router).

The SBC is installed in a VM on the NAS (synology).

 

[NickD_3CX]

We tried with the firewall by opening all ports but that did not help but we don't have access to the router.

On the router between the SBC and the internet, there don't need to be any ports open Inbound, you only need to make sure the SBC can make outbound connections freely. I'd even suggest closing those ports if you have opened them.

The SBC is installed in a VM on the NAS (synology).

This is not really supported. I know the Synology NAS (I own one myself) has the "Virtual Machine Manager" that allows you to spin up VMs, but the rest of the operations may be taking up too many resources at these times. Plus, I don't know what virtual network you have set it up to use.

Also, having 40 extension behnind it requires quite a bit of resources, so you may have not allocated enough.

Anyhow, I'd highly recommend you look into aquiring a Raspberry Pi and installing the SBC on there. The cost is not that big and you would for sure rule out any problems due to the NAS.

 

[litus]

Thank you so far by helping us.

We will install SBC on Raspberry Pi and look forward to solve the problem.

Kind regards,
Marius