[HCoop-Discuss] HCoop maintenance this Sunday (user daemon restart required)

[Richard Darst]

Hello,

As part of the ongoing migration, there will be a bit of user-visible
maintenance this Sunday.


As part of the follow-up to the security incidents earlier this year,
we will regenerate the master kerberos key and all user daemon keys.
When this happens, all services will have to restart.  We will handle
all HCoop services, however if you have daemons running you will have
to restart them yourself.


Also, to increase overall HCoop performance, we will be upgrading our
SQL database software and moving it to the new server.

All your databases (if you host any at HCoop) should be transferred
automatically and require no work on your part.

For database upgrades, please verify that your services use hosts
mysql.hcoop.net or postgres.hcoop.net instead of any specific machine
names.  After our maintenance work, please check whether any of your
custom services require a restart to pick up the changes in case the
old servers are cached.


This upgrade is planned for Sunday, March 21.  The maintenance will
take place between 11UTC and 22UTC, over a yet-undecided hour or two
block.  Estimated downtime will be no more than few hours.  For
information as to the exact time of maintenance, please see the
identica hcoop status group below.

Should you notice any problems, please submit a support request
through our Bugzilla interface at
  https://bugzilla.hcoop.net/

Status updates will be avialable at:
  http://identi.ca/group/hcoopstatus
which is also a good place to look for any other hcoop service status
updates.

-- 
Thanks,
Richard Darst on behalf of HCoop
#hcoop on freenode
Emergency status updates: http://identi.ca/group/hcoopstatus