neutron

neutron takes long to reconnect to the db

Bug #1298310 reported by Rossella Sblendido
8
This bug affects 1 person
Affects Status Importance Assigned to Milestone
neutron
New
Undecided
Dirk Mueller

Bug Description

In an HA setup, DB node 1 goes down and the DB is fail-overing to DB node 2 using the same IP. Neutron reports

OperationalError: (OperationalError) socket not open

This is because the TCP connection is broken.
Anyway Neutron takes minutes to re-connect to the new DB node.

Revision history for this message
Dirk Mueller (dmllr) wrote :

This is actually a duplicate of https://bugs.launchpad.net/fuel/+bug/1288438 but for postgreSQL.

Fix is here: https://review.openstack.org/#/c/85980/ (needed on top of https://review.openstack.org/#/c/81225/)

Changed in neutron:
assignee: nobody → Dirk Mueller (dmllr)
tags: added: ha
tags: added: icehouse-backport-potential icehouse-rc-potential
Thierry Carrez (ttx)
tags: removed: icehouse-rc-potential
Revision history for this message
Salvatore Orlando (salvatore-orlando) wrote :

Thanks I will then backport the patches.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.
Subscribing...

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.