How to troubleshoot "write on backend 0 failed"

From: "David O'Mahony" <david(dot)c(dot)omahony(at)gmail(dot)com>
To: pgsql-general(at)postgresql(dot)org
Subject: How to troubleshoot "write on backend 0 failed"
Date: 2017-02-09 20:46:44
Message-ID: CAD93rxr-FDG8CxMw8msLA7wo1au8CrLsB_=g3KfJ1hJ7XD6CjQ@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Hi All,

We are using pgpool 3.6.1 with two nodes in the cluster. We are seeing
frequent occurrences where one node will be marked as offline.

When this occurs we are seeing the following statements in our logs.

2017-02-09 09:45:38: pid 12125: WARNING: write on backend 0 failed with
error :"Success"
2017-02-09 09:45:38: pid 12125: DETAIL: while trying to write data from
offset: 0 wlen: 5

How do we troubleshoot the issue?

From looking at the source, the message appears to be generated from
pool_stream.c:pool_flush_it().

Based on the fact that the strerr() returns "Success", did anything
actually go wrong? What if the write() function simply returned 0 bytes
written? That's not necessarily an error is it?

Thanks in advance for your help,

Dave

Browse pgsql-general by date

  From Date Subject
Next Message Joshua Chamberlain 2017-02-09 21:56:33 CREATE TABLE with parallel workers, 10.0?
Previous Message Rakesh Kumar 2017-02-09 18:27:30 Re: clarification about standby promotion