0001785: When database is stoped, pfmon ends in an inconsistent state - PacketFence

Bug Tracking System

View Issue Details [ Jump to Notes ]

[ Issue History ] [ Print ]

Project

Category

View Status

Date Submitted

Last Update

0001785

PacketFence

core

public

2014-04-04 08:04

Reporter

canepan

Assigned To

Priority

normal

Severity

major

Reproducibility

always

Status

new

Resolution

open

Platform

Linux

RHEL / CentOS

OS Version

Product Version

4.1.0

Target Version

Fixed in Version

Summary

0001785: When database is stoped, pfmon ends in an inconsistent state

Description

To perform backup, we stop the database of PacketFence every night.
If pfmon tries to access the DB during the backup, it logs:
Apr 02 18:00:23 pfmon(0) INFO: running expire check (main::cleanup)
Apr 02 18:00:23 pfmon(0) INFO: checking registered nodes for expiration (main::cleanup)
Apr 02 18:00:23 pfmon(0) WARN: database query failed with: MySQL server has gone away. (errno: 2006), will try again (pf::db::db_query_execute)
Apr 02 18:00:23 pfmon(0) FATAL: unable to connect to database: Can't connect to MySQL server on '127.0.0.1' (111) at /usr/local/pf/lib/pf/node.pm line 888.
(pf::db::db_connect)
Apr 02 18:00:23 pfmon(0) ERROR: Error restarting pfmon: unable to connect to database: Can't connect to MySQL server on '127.0.0.1' (111) at /usr/local/pf/lib/pf/node.pm line 888.
(main::cleanup)

Also, pfmon results running with "ps", but "pfcmd service pfmon status" doesn't agree:
/usr/local/pf # ./bin/pfcmd service pfmon status
service|shouldBeStarted|pid
pfmon|1|0

so it's not possible to restart it without doing kill to the running process.

Nodes expired after this happening are not requested login again (they are registered forever), and so they do not expire

Steps To Reproduce

1) start PacketFence
2) stop database
3) wait a minute

Issue History
Date Modified	Username	Field	Change
2014-04-04 08:04	canepan	New Issue

Relationships

Notes
There are no notes attached to this issue.