View Issue Details
|ID||Project||Category||View Status||Date Submitted||Last Update|
|0000392||Pgpool-II||Bug||public||2018-04-16 15:03||2019-01-30 10:05|
|Target Version||Fixed in Version|
|Summary||0000392: PGpool master lost connection to database even if the other services and pgpool slave didn't lost connecton.|
Yesterday at one of our costumer server we had strange situation.
PGpool master ( with virtual address ) lost connection to main database even if the other applications and services connected directly to DB dind't lost connection. It means that the network connection was good. Also diagrams with network connections looks good.
• First time PGpool master lost connection at 10:20
"10:20:22: pid 58531: ERROR: unable to read data from DB node 0"
• Next one was at 10:46 next one at 10:47:23 and then pgpool started write into log file the "check health" status.
"10:47:23: pid 58531: LOG: health checking retry count 3"
• At 10:50:10 reached max retries times and made Failover file to stop replication, then pgpool started write to log that "can not accept new connections".
"2018-04-12 10:50:10: pid 58531: LOG: setting backend node 0 status to NODE DOWN"
• At 11:03:39 we had to restart PGpool and attach db main to pgpool
• After this time pgpool works good again .
We don't know why the pgpool lost connection even if other applications connected directly to DB didn't lost connection. We think that there is some issues with config file. Can you check this and help us to resolve this problem? This situation occure from time to time, but it is very onerous due to the fact that we do not know what is happening and what is the reason. in the attachment I attach files with logs and configuration.
|Tags||No tags attached.|