<div dir="ltr">Now expect those two lines:<div><br></div><div><div>pgpool1[24660]: [13034-1] 2015-04-15 10:00:10: pid 24660: FATAL: unable to read data from frontend</div><div>pgpool1[24660]: [13034-2] 2015-04-15 10:00:10: pid 24660: DETAIL: EOF encountered with frontend</div></div><div><br></div><div>I have nothing in log, even failover is not triggered. </div><div>Cluster was ok for some time (about 20 hours), but when traffic is increased, we gave degeneration... any hinting where I can look at?</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Apr 13, 2015 at 10:30 AM, Armin Nesiren <span dir="ltr"><<a href="mailto:armin.nesiren@intera.si" target="_blank">armin.nesiren@intera.si</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Hello,<div><br></div><div>I'm running pgpool + two postgresql servers in streaming replication with load balancing enabled.</div><div><br></div><div>pgpool version is 3.4.1 (tataraboshi)<br clear="all"><div>postgresql is 9.3</div><div><br></div><div>Recently I have a lot of failovers that is happening for unknown reason to me and puts my master node out. </div><div><br></div><div>This is part of my log that is interesting, this is when failover is started:</div><div><br></div><div><div><br></div><div><font face="monospace, monospace">Apr 13 09:20:28 localhost pgpool1[28256]: [3299-1] 2015-04-13 09:20:28: pid 28256: LOG: pool_send_and_wait: Error or notice message from backend: : DB node id: 1 backend pid: 6667 statement: "begin;" message: "there is already a transaction in progress"</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">Apr 13 09:20:31 localhost pgpool1[4539]: [794-1] 2015-04-13 09:20:31: pid 4539: LOG: received degenerate backend request for node_id: 0 from pid [4539]</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace"><b>Apr 13 09:20:31 localhost pgpool1[10204]: [3171-1] 2015-04-13 09:20:31: pid 10204: LOG: starting degeneration. shutdown host node1(5432)</b></font></div><div><font face="monospace, monospace"><b><br></b></font></div><div><font face="monospace, monospace"><b>Apr 13 09:20:31 localhost pgpool1[10204]: [3172-1] 2015-04-13 09:20:31: pid 10204: LOG: Restart all children</b></font></div><div><font face="monospace, monospace"><b><br></b></font></div><div><font face="monospace, monospace"><b>Apr 13 09:20:31 localhost pgpool1[10204]: [3173-1] 2015-04-13 09:20:31: pid 10204: LOG: execute command: /etc/pgpool1/failover.sh 0 "node1" 5432 /var/lib/pgsql/9.3/data 1 0 "node2" 0</b></font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">Apr 13 09:20:31 localhost pgpool1[23945]: [2964-1] 2015-04-13 09:20:31: pid 23945: LOG: child process received shutdown request signal 3</font></div><div><font face="monospace, monospace"><br></font></div><div><font face="monospace, monospace">Apr 13 09:20:31 localhost pgpool1[28278]: [3166-1] 2015-04-13 09:20:31: pid 28278: LOG: child process received shutdown request signal 3</font></div></div><div><br></div><div>Part of pgpool configuration that might be interesting:</div><div><br></div><div><div><font face="monospace, monospace">replication_stop_on_mismatch = off</font></div><div><font face="monospace, monospace">failover_if_affected_tuples_mismatch = off</font></div></div><div><div><font face="monospace, monospace">health_check_period = 30</font></div><div><font face="monospace, monospace">health_check_timeout = 60</font></div></div><div><br></div><div><br></div><div>Of course I can provide more info if needed.</div><div>Can someone please help me where to look and how to indentify/prevent this problem.</div><div><br></div><div>Thanks a lot. </div><span class="HOEnZb"><font color="#888888"><div><br></div>-- <br><div>Armin</div>
</font></span></div></div>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature">Armin</div>
</div>