Backend cache problem revisited
От | Yin-So Chen |
---|---|
Тема | Backend cache problem revisited |
Дата | |
Msg-id | 37AD1173.CFB64025@uswest.net обсуждение исходный текст |
Список | pgsql-general |
Hi, after reading through the archives it seems that this particular error NOTICE: SIAssignBackendId: discarding tag 2147483553 Connection to database failed FATAL 1: Backend cache invalidation initialization failed had not come up for a while for most people. However I didn't see any postings about a solution either, so I tried to figure out if this problem went away after the 6.3.2 build. On 6.3.2 this error occurs after there are more than 32 concurrent connections. On 6.4.2 the number of concurrent connections allowed doubled to 64, but it occurs again if you create more. My system environment has dual PII450, 1GB RAM, and I am running RedHat5.2 with Apache1.3.6 & mod_perl & Postgresql6.3.2 (all RedHat builds except for the kernel). I use mod_perl to keep the database connection persistent. With 1GB of ram I can hold at least 100 connections open at the same time, and I would really like to be able to utilize my computer's resources fully. Can anyone tell me how to reconfigure postgresql so I can avoid this error when opening more connections? Any help is greatly appreciated. Regards, Yin-So Chen
В списке pgsql-general по дате отправления: