could not fork new process and out of memory issues PG 11.5 - Mailing list pgsql-bugs

From Zhiyu ZY13 Xu
Subject could not fork new process and out of memory issues PG 11.5
Date
Msg-id HK2PR03MB4610CA0D5B9B9F9424DC15A7A8DA0@HK2PR03MB4610.apcprd03.prod.outlook.com
Whole thread Raw
List pgsql-bugs

Hi Support

 

I experience a "out of memory and could not fork new process " problem on PG 11.5. This problem occur sometimes. On that time the workload not very high.

 Linux system have enough free memory. New connection unable to connect to database.

Some connected connection will throw memeory not enough releated errors.  I am not clearly what problem caused this issue. I feel I have allocate enough resource to this system.

 Would you like to help me narow down this issue ?  I attach the logs and database configuration for your reference. Thanks   

 

 Env:

PG 11.5

 Linux Centos 7.5

 Memory: 8G

 Totally DB size: 700MB

 

Errors:

 

2020-12-26 21:00:47.141 CST,"zabbix","postgres",24739,"127.0.0.1:48814",5fe733ff.60a3,3,"SELECT",2020-12-26 21:00:47 CST,19/442688,0,ERROR,53200,"out of memory","Failed on request of size 200 in memory context ""MessageContext"".",,,,,"select count(*) AS server_connections from pg_stat_activity where backend_type = 'client backend';",,,"psql"

2020-12-26 21:00:47.144 CST,"zabbix","postgres",24739,"127.0.0.1:48814",5fe733ff.60a3,4,"SELECT",2020-12-26 21:00:47 CST,19/442689,0,ERROR,53200,"out of memory","Failed on request of size 16 in memory context ""MessageContext"".",,,,,"select count(*) AS locks_waiting from pg_stat_activity where backend_type = 'client backend' and wait_event_type like '%Lock%';",,,"psql"

 

2020-12-27 00:28:01.520 CST,,,63552,,5facd1fc.f840,9,,2020-11-12 14:11:08 CST,,0,LOG,00000,"could not fork autovacuum worker process: Cannot allocate memory",,,,,,,,,""

2020-12-27 00:28:02.521 CST,,,63552,,5facd1fc.f840,10,,2020-11-12 14:11:08 CST,,0,LOG,00000,"could not fork autovacuum worker process: Cannot allocate memory",,,,,,,,,""

2020-12-27 00:28:03.522 CST,,,63552,,5facd1fc.f840,11,,2020-11-12 14:11:08 CST,,0,LOG,00000,"could not fork autovacuum worker process: Cannot allocate memory",,,,,,,,,""

2020-12-27 00:28:04.524 CST,,,63552,,5facd1fc.f840,12,,2020-11-12 14:11:08 CST,,0,LOG,00000,"could not fork autovacuum worker process: Cannot allocate memory",,,,,,,,,""

2020-12-27 00:28:05.525 CST,,,63552,,5facd1fc.f840,13,,2020-11-12 14:11:08 CST,,0,LOG,00000,"could not fork autovacuum worker process: Cannot allocate memory",,,,,,,,,""

2020-12-27 00:28:06.526 CST,,,63552,,5facd1fc.f840,14,,2020-11-12 14:11:08 CST,,0,LOG,00000,"could not fork autovacuum worker process: Cannot allocate memory",,,,,,,,,""

2020-12-27 00:28:07.527 CST,,,63552,,5facd1fc.f840,15,,2020-11-12 14:11:08 CST,,0,LOG,00000,"could not fork autovacuum worker process: Cannot allocate memory",,,,,,,,,""

2020-12-27 00:28:08.528 CST,,,63552,,5facd1fc.f840,16,,2020-11-12 14:11:08 CST,,0,LOG,00000,"could not fork autovacuum worker process: Cannot allocate memory",,,,,,,,,""

2020-12-27 00:28:09.007 CST,,,63552,,5facd1fc.f840,17,,2020-11-12 14:11:08 CST,,0,LOG,00000,"could not fork new process for connection: Cannot allocate memory",,,,,,,,,""

 

徐志宇(Jack

Database Engineer

    

DB Team,ITS. Lenovo China

Phone: 86-18910860709

Email:xuzy13@lenovo.com

No.6 Shangdi West Road, Haidian District Beijing, China, 100085

 

Attachment

pgsql-bugs by date:

Previous
From: Jeff Davis
Date:
Subject: Re: BUG #16784: Server crash in ExecReScanAgg()
Next
From: PG Bug reporting form
Date:
Subject: BUG #16792: silent corruption of GIN index resulting in SELECTs returning non-matching rows