Daniel-Constantin Mierla writes:
Are the two connections going to the same mysql
server? Or different ones?
Different ones.
Are you able to see from the logs if writing to the
first connection
failed and it tried to the second one?
No, everything seems normal. Here is the last working acc insert before the freeze:
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: acc [acc_logic.c:446]:
should_acc_reply(): acc is on
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: db_cluster [dbcl_api.c:482]:
db_cluster_use_table(): use table (accounting) - cluster [sip_proxy_usage_cluster]
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: db_cluster [dbcl_api.c:491]:
db_cluster_use_table(): set read table (accounting) - cluster [sip_proxy_usage_cluster]
(5/0)
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: db_cluster [dbcl_api.c:501]:
db_cluster_use_table(): set write table (accounting) - cluster [sip_proxy_usage_cluster]
(5/0)
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: db_cluster [dbcl_api.c:491]:
db_cluster_use_table(): set read table (accounting) - cluster [sip_proxy_usage_cluster]
(4/0)
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: db_cluster [dbcl_api.c:501]:
db_cluster_use_table(): set write table (accounting) - cluster [sip_proxy_usage_cluster]
(4/0)
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: db_cluster [dbcl_api.c:378]:
db_cluster_insert(): executing db cluster insert command
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: db_cluster [dbcl_api.c:379]:
db_cluster_insert(): serial operation - cluster [sip_proxy_usage_cluster] (5/0)
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: acc [acc_extra.h:63]:
free_strar_mem(): Freeing memory, type is 2, message_index 7, index i 0
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: tm [t_reply.c:1667]:
cleanup_uac_timers(): RETR/FR timers reset
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: <core> [core/receive.c:442]:
receive_msg(): reply-route executed in: 1395 usec
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: <core> [core/usr_avp.c:636]:
destroy_avp_list(): destroying list (nil)
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: <core> [core/usr_avp.c:636]:
destroy_avp_list(): destroying list (nil)
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: <core> [core/usr_avp.c:636]:
destroy_avp_list(): destroying list (nil)
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: <core> [core/usr_avp.c:636]:
destroy_avp_list(): destroying list (nil)
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: <core> [core/usr_avp.c:636]:
destroy_avp_list(): destroying list (nil)
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: <core> [core/usr_avp.c:636]:
destroy_avp_list(): destroying list (nil)
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: <core> [core/xavp.c:495]:
xavp_destroy_list(): destroying xavp list (nil)
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: <core> [core/receive.c:458]:
receive_msg(): cleaning up
Sep 2 12:00:20 rox2 /usr/bin/sip-proxy[341]: DEBUG: <core> [core/udp_server.c:491]:
udp_rcv_loop(): received on udp socket: (112/100/1063) [[INVITE ...]]
After that there is a couple of these:
Sep 2 12:00:21 rox2 /usr/bin/sip-proxy[341]: DEBUG: acc [acc_logic.c:413]:
should_acc_reply(): probing acc state - code: 183 flags: 0x200209c0
Sep 2 12:00:21 rox2 /usr/bin/sip-proxy[341]: DEBUG: acc [acc_logic.c:442]:
should_acc_reply(): early media acc is off
And finally the last should_acc_reply() before the freeze that does not
result to any kind db_cluster entry in syslog:
Sep 2 12:00:21 rox2 /usr/bin/sip-proxy[341]: DEBUG: acc [acc_logic.c:413]:
should_acc_reply(): probing acc state - code: 480 flags: 0x240005c4
Sep 2 12:00:21 rox2 /usr/bin/sip-proxy[341]: DEBUG: acc [acc_logic.c:430]:
should_acc_reply(): failed acc is on
Sep 2 12:00:21 rox2 /usr/bin/sip-proxy[341]: DEBUG: tm [t_lookup.c:1111]: t_check_msg():
msg (0x7f2383bea9e0) id=12872/341 global id=12872/341 T end=0x7f236b1cc3f0
Sep 2 12:00:21 rox2 /usr/bin/sip-proxy[341]: DEBUG: tm [t_reply.c:2243]:
reply_received(): transaction found - T:0x7f236b1cc3f0 branch:0
Next process 341 entry in syslog is 13+ minutes later after the freeze:
Sep 2 12:13:55 rox2 /usr/bin/sip-proxy[341]: DEBUG: tm [t_reply.c:2256]:
reply_received(): org. status uas=408, uac[0]=408 local=0 is_invite=1)
-- Juha