1

I recently upgrade a VM to Redhat 5 and moved Sphinx Search to that server. For some reason when I try to start Sphinx, the startup script fails to push the searchd process to the background. I created a test script with a minimal version of the startup call, which is as follows:

#!/bin/sh
. /etc/rc.d/init.d/functions
daemon /usr/local/sphinx/bin/searchd --config
/home/app/code/server_config/sphinx_config.php

The searchd process starts and I get output that indicates it is listening on the correct IP:port, but the process never returns.

The process that is hanging is in the daemon call, and ps shows it as:

/bin/bash -c ulimit -S -c 0 >/dev/null 2>&1 ; /usr/local/sphinx/bin/searchd --config
/home/app/code/server_config/sphinx_config.php

Any ideas on what might be causing this or how to troubleshoot? We use the same process on about a dozen other servers running Redhat 4 which work fine (although the process created by daemon uses initlog, so it is different).

Jon St. John
  • 276
  • 2
  • 4

1 Answers1

0

Turns out that there was nothing wrong with the init script, but that sphinx searchd was taking a very long time to start. Insufficient memory was allocated on the VM, so it was having problems starting up with the large indexes. Increased the memory allocation and works fine now.

Jon St. John
  • 276
  • 2
  • 4