Opened 10 months ago

Last modified 9 months ago

#127 new defect

1984.is Server Rebuild

Reported by: chris Owned by: chris
Priority: major Milestone: Maintenance
Component: crin1 Version:
Keywords: Cc: peter
Estimated Number of Hours: 0 Add Hours to Ticket: 0
Billable?: yes Total Hours: 9.63

Description

Due to critical hardware failure at 1984.is all three servers needed rebuilding from scratch.

Change History (8)

comment:1 Changed 10 months ago by chris

  • Add Hours to Ticket changed from 0 to 5.0
  • Total Hours set to 5

comment:2 Changed 10 months ago by chris

  • Add Hours to Ticket changed from 0 to 0.63
  • Total Hours changed from 5 to 5.63

Email isn't working for Trac but Munin and Piwik are now up:

And the https://crin.org/ is also up but there might be missing PHP modules and things like Solr -- some testing will be needed.

Last edited 10 months ago by chris (previous) (diff)

comment:3 Changed 10 months ago by chris

  • Add Hours to Ticket changed from 0 to 0.25
  • Total Hours changed from 5.63 to 5.88

phpmyadmin is up and running:

comment:4 Changed 10 months ago by chris

  • Add Hours to Ticket changed from 0 to 0.3
  • Total Hours changed from 5.88 to 6.18

Outgoing email might be working now, Exim has been installed and the config copied over.

There was a mistake in the backups, /usr/local wasn't backed up so various scripts (which are on this Trac site) and Munin plugins will need installing from scratch.

Non-root users haven't been added yet and the nothing has been done to get the dev server up and running.

comment:5 Changed 10 months ago by chris

  • Add Hours to Ticket changed from 0 to 2
  • Total Hours changed from 6.18 to 8.18

Missing WikiStart#Scripts have been installed on the servers, several using the versions in this repo.

User accounts have been added — people should be able to ssh in again.

The missing Munin plugins have been sorted out and it is possible that Trac email is working...

comment:6 Changed 10 months ago by chris

  • Add Hours to Ticket changed from 0 to 0.65
  • Cc peter added
  • Total Hours changed from 8.18 to 8.83

Checking the /var/log/exim4/mainlog log confirms that the comment above did result in a email.

I have added Peter to this ticket in case there are any issues he wants to flag up.

I'll get the dev server up and running again in December.

comment:7 Changed 9 months ago by chris

  • Add Hours to Ticket changed from 0 to 0.3
  • Total Hours changed from 8.83 to 9.13

There is quite a mailq on crin2, looking at the install notes, ticket:6#comment:18, dpkg-reconfigure exim4-config was run and /etc/aliases was edited -- it looks like all the emails are undelivered email to root, so exim -qff was run to flush the queue.

Also /etc/exim4/exim4.conf.localmacros was updated, and update-exim4.conf was run and a test email was sent to root and everything seems to be working OK now.

comment:8 Changed 9 months ago by chris

  • Add Hours to Ticket changed from 0 to 0.5
  • Total Hours changed from 9.13 to 9.63

One of the, now working, emails:

PHP Warning:  PHP Startup: Unable to load dynamic library '/usr/lib/php5/20131226/uploadprogress.so' - /usr/lib/php5/20131226/uploadprogress.so: cannot open shared object file: No such file or directory in Unknown on line 0                                                                                                                                         
PHP Warning:  PHP Startup: Unable to load dynamic library '/usr/lib/php5/20131226/memcached.so' - /usr/lib/php5/20131226/memcached.so: cannot open shared object file: No such file or directory in Unknown on line 0
PHP Warning:  PHP Startup: Unable to load dynamic library '/usr/lib/php5/20131226/newrelic.so' - /usr/lib/php5/20131226/newrelic.so: cannot open shared object file: No such file or directory in Unknown on line 0

So following the install notes, ticket:6#comment:15 and ticket:117

pecl install uploadprogress
aptitude install php5-memcached
echo "deb https://apt.newrelic.com/debian/ newrelic non-free" > /etc/apt/sources.list.d/newrelic.list
wget -O- https://download.newrelic.com/548C16BF.gpg | sudo apt-key add -
apt-get update
apt-get install newrelic-php5 newrelic-sysmond
service php5-fpm restart

Then the New Relic key was copied for the backups and added to /etc/newrelic/nrsysmond.cfg and then it was started, /etc/init.d/newrelic-sysmond start.

I also whitelisted munijn.webarch.net so we can have some graphs to compare with the Iceland hosting when that is back online -- the load issues we had there with PHP seem to have gone away, see:

https://munin.crin.org/munin/crin.org/crin2.crin.org/#php

Note: See TracTickets for help on using tickets.