[OpenSIPS-Users] Usrloc can't be sync after upgrade to 2.4.8

[William Jin]

Found some extra error logs that might be related to this. checking how we can fix this

/usr/sbin/opensips[1983]: ERROR:proto_bin:add_write_chunk: We have reached the limit of max async postponed chunks
/usr/sbin/opensips[1983]: ERROR:proto_bin:proto_bin_send: failed to send
/usr/sbin/opensips[1983]: ERROR:clusterer:msg_send: send() to 10.2.2.16:5566 for proto bin/7 failed


--
Regards,
William Jin
________________________________
From: William Jin <williamj at exetel.com.au>
Sent: Thursday, 2 July 2020 11:24 AM
To: OpenSIPS users mailling list <users at lists.opensips.org>
Subject: Re: [OpenSIPS-Users] Usrloc can't be sync after upgrade to 2.4.8

Hi,

Thanks for the info.

I have upgraded both nodes to latest 2.4.8, and restarted opensips on both nodes so it should start from empty.
The error "INFO:usrloc:receive_binary_packets: discarding , ver 1: need ver 2" is gone.

However, the 2nd node usrloc is still not synced with 1st node(seed), even manually issue the command opensipsctl fifo ul_cluster_sync.

my config: (on both node)
# ----- usrloc params -----
modparam("usrloc", "cluster_mode", "full-sharing")
modparam("usrloc", "nat_bflag", "NAT")
modparam("usrloc", "location_cluster", 1)
modparam("usrloc", "restart_persistency", "sync-from-cluster")

The clusterer table:
 id | cluster_id | node_id |        url         | state | no_ping_retries | priority | sip_addr | description | flags
----+------------+---------+--------------------+-------+-----------------+----------+----------+-------------+-------
  2 |          1 |       2 | bin:10.2.2.16:5566 |     1 |               3 |       50 |          | node2     |
  1 |          1 |       1 | bin:10.2.2.15:5566 |     1 |               3 |       50 |          | node1      | seed

>From node1 log:
Jul  2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:handle_internal_msg: Node [2] is UP
Jul  2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:handle_sync_request: Received sync request for capability 'dialog-dlg-repl' from node 2, cluster 1
Jul  2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:send_sync_repl: Sent all sync packets for capability 'dialog-dlg-repl' to node 2, cluster 1
Jul  2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:handle_sync_request: Received sync request for capability 'usrloc-contact-repl' from node 2, cluster 1
Jul  2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:msg_send: send() to 10.2.2.16:5566 for proto bin/7 failed
Jul  2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:msg_send_retry: msg_send() to node [2] failed
Jul  2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:cl_sync_chunk_start: Failed to send sync packet
...
Jul  2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:cl_sync_chunk_start: Failed to send sync packet
Jul  2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:send_sync_repl: Failed to send sync packet, rc=-1
Jul  2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:send_sync_repl: Failed to send sync end message
Jul  2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:handle_internal_msg: Node [2] is UP

Node2 log:
Jul  2 10:55:56 /usr/sbin/opensips[13782]: INFO:clusterer:handle_internal_msg: Node [1] is UP
Jul  2 10:55:56 /usr/sbin/opensips[13782]: INFO:clusterer:send_sync_req: Sent sync request for capability 'dialog-dlg-repl' to node 1, cluster 1
Jul  2 10:55:56 /usr/sbin/opensips[13782]: INFO:clusterer:send_sync_req: Sent sync request for capability 'usrloc-contact-repl' to node 1, cluster 1
Jul  2 10:55:56 /usr/sbin/opensips[13782]: INFO:clusterer:handle_sync_packet: Received all sync packets for capability 'dialog-dlg-repl' in cluster 1
Jul  2 10:55:56 /usr/sbin/opensips[13793]: INFO:core:probe_max_sock_buff: using snd buffer of 512 kb
Jul  2 10:55:56 /usr/sbin/opensips[13793]: INFO:core:init_sock_keepalive: TCP keepalive enabled on socket 184

we have around 30,000 end devices(UDP), and I found the https://github.com/OpenSIPS/opensips/issues/1976, not sure if it's related

Any idea? Do I need to increase the ping_timeout and give it a try?

--
Regards,
William Jin
________________________________
From: Users <users-bounces at lists.opensips.org> on behalf of Liviu Chircu <liviu at opensips.org>
Sent: Wednesday, 1 July 2020 4:48 PM
To: OpenSIPS users mailling list <users at lists.opensips.org>
Subject: Re: [OpenSIPS-Users] Usrloc can't be sync after upgrade to 2.4.8

On 01.07.2020 09:39, Liviu Chircu wrote:
> then re-register them manually with the "ul_add" [2] command that and
> some Python scripting logic, for example.

Sorry, I meant bash, not Python.  Not that I have anything against
Python, but the former script (basically a simple loop) is much quicker
to write in order to achieve your purpose.

--
Liviu Chircu
www.twitter.com/liviuchircu<http://www.twitter.com/liviuchircu> | www.opensips-solutions.com<http://www.opensips-solutions.com>


_______________________________________________
Users mailing list
Users at lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/users/attachments/20200702/249a7a64/attachment.html>