Commit Graph

265 Commits

Author SHA1 Message Date
JefftheBest1 9de75f5167 Fixed typos with separate 2017-01-12 04:52:05 -08:00
JefftheBest1 ff8639f9db Fixed typos with threshold. 2017-01-12 04:50:20 -08:00
JefftheBest1 5008f45112 Fixed typo in munin.conf.erb 2017-01-12 04:49:19 -08:00
Tim Abbott 3e32102016 nagios: Fix various critical issues not tagged as pageable. 2017-01-06 21:49:20 -08:00
Tim Abbott edebf7619b puppet: Add PAM common_session disabling systemd-login.
This fixes a weird problem with systemd where logging into a server
via ssh frequently has a 15s+ lag.
2017-01-06 21:49:15 -08:00
Tim Abbott 93c2c19775 nagios: Increase process count limits. 2017-01-06 21:49:15 -08:00
Tim Abbott 2c6cb37385 munin: Add default munin configuration template. 2017-01-06 21:44:57 -08:00
Tim Abbott 9ab8e7ba34 nagios: Disable swap checks for servers with no swap. 2017-01-06 21:39:07 -08:00
Tim Abbott 3e01ed1f73 nagios: Increase NTP max_check_attempts.
NTP often suffers from brief interruptions of service that lead to
spurious Nagios alerts; it makes sense to suppress these.
2017-01-06 21:32:43 -08:00
Tim Abbott e4420b08d2 zulip_ops: Disable unattended upgrades of security packages.
Since Zulip does not handle e.g. postgres server restarts gracefully,
it's best for a system administrator to manually trigger security
updates.
2017-01-06 21:30:56 -08:00
Tim Abbott 6f9c73d0e5 zmirror: Update Debathena release in configuration.
The zulip_ops configuration is now for xenial, not obsolete wheezy.
2017-01-06 21:30:41 -08:00
Tim Abbott bd9176d1d9 nagios: Remove some default files.
Nagios ships with a bunch of default configuration files that one
needs to delete in order to configure it.
2017-01-06 21:25:12 -08:00
Tim Abbott 7083899e77 zulip_ops: Add postgres config for enabling Nagios.
The old zulip_ops Nagios configuration depended on Nagios having the
ability to login as the zulip user (with essentially full write
access); this configuration is helpful for limiting nagios to special
"nagios" user with more limited credentials.
2017-01-06 21:24:24 -08:00
Tim Abbott 204edb0f85 zulip_ops: Cleanup pg_hba.conf configuration. 2017-01-06 21:23:51 -08:00
Tim Abbott 30c57eb2ae zulip_ops: Add basic .emacs for production. 2017-01-06 21:20:21 -08:00
Tim Abbott eb87d04168 puppet: Remove xxxxx password hardcoding in recovery.conf. 2017-01-06 21:20:21 -08:00
Tim Abbott 6404a1a5ff zulip_ops: Add nagios-plugins-contrib.
This has a number of useful nagios plugins.
2017-01-06 21:19:59 -08:00
Tim Abbott f7b77008ef zulip_ops: Add aptitude dependency.
This is useful for `aptitude why`.
2017-01-06 21:19:50 -08:00
Tim Abbott 2510a51a8a zulip_ops: Add letsencrypt dependency. 2017-01-06 21:19:31 -08:00
Tim Abbott 65774e1c4f zulip_ops: use check_postgres package from apt. 2017-01-06 21:18:55 -08:00
K.Kanakhin 0d8c18a6dd nagios-plugins: Add websocket checking to nagios message sending test.
- Add websocket client to create connection with SockJS websocket server.
  It contains callback method to launch after connection setup.
- Add '--websocket' parameter to 'check_send_receive_time' script to
  check websocket connection.
- Add testing  websocket connection to production installation checking.
- Add cronjob to launch websocket connection nagios test.

This makes it possible for Zulip Nagios monitoring to check for
problems impacting the websockets sending code path, which is what all
web users use.
2016-12-30 15:36:37 -08:00
Jason Le 144d82305d mypy: Annotate puppet/zulip_ops. 2016-12-03 11:00:25 -08:00
bulat22101 adebc75740 pep8: Fix E502 violations 2016-12-03 10:56:36 -08:00
Sidhant Bhavnani 8c0c12c1d9 pep8: Fix E303 violations. 2016-12-02 15:34:11 -08:00
Rafid Aslam c5316b4002 lint: Fix E127 pep8 violations.
Fix pep8: E127 continuation line over-indented for visual indent
style issue.
2016-12-01 10:23:55 -08:00
Rafid Aslam 7a2282986a pep8: Fix E225 pep8 violations. 2016-11-28 15:21:15 -08:00
Anders Kaseorg 207cf6302b Always start python via shebang lines.
This is preparation for supporting using Python 3 in production.

Signed-off-by: Anders Kaseorg <andersk@mit.edu>
2016-11-26 14:46:37 -08:00
Tim Abbott 2e65dc1206 puppet: make check_send_receive_time target host configurable. 2016-11-02 23:40:53 -07:00
Tim Abbott eceaf36001 setup_disks: Fix postgres RAID setup to work correctly on Xenial.
Nobody's going to run this on Wheezy again.
2016-10-28 11:04:08 -07:00
Tim Abbott 9b7a3f040c Remove now-unused /json/get_events endpoint. 2016-10-27 21:34:58 -07:00
Tim Abbott 4fbe201187 puppet: Automate autossh process monitoring maintenance.
Previously, the Zulip Nagios configuration effectively hardcoded the
count for how many system should have autossh connections.
2016-10-26 00:49:03 -07:00
Tim Abbott 6bdb10b71b puppet: Update emacs dependency to emacs-nox metapackage.
This way, one doesn't need to keep updating the dependency every time
a new major emacs release comes out.
2016-10-26 00:42:22 -07:00
Tim Abbott 11b5d203f7 sshd_config: Increase MaxStartups.
This fixes connection problems when using the full Zulip recommended
Nagios configuration against a given server.
2016-10-26 00:41:03 -07:00
Tim Abbott 73f54dd0cb sshd_config: Add updates from Xenial upstream.
It seems worth updating this to match the Linux distro this
configuration targets.
2016-10-26 00:40:44 -07:00
Tim Abbott 0a5a2c4eda nagios: Automate authorized users list maintenance. 2016-10-26 00:37:29 -07:00
Tim Abbott fa4998db59 puppet: Add zulip_zephyr_mirror plugins. 2016-10-26 00:35:57 -07:00
Tim Abbott ac4f28050c zmirror: Remove unnecessary krb5-clients dependency.
I'm pretty sure krb5-clients isn't needed to run the Zephyr mirroring
service.
2016-10-26 00:35:11 -07:00
Tim Abbott d490e83645 puppet: Upgrade nagios cgi.cfg with modern defaults. 2016-10-26 00:31:41 -07:00
Tim Abbott 1159ad4857 puppet: Upgrade nagios.cfg with modern defaults. 2016-10-26 00:31:41 -07:00
Tim Abbott 73178e5e5a puppet: Run check_send_receive_time via a cron job.
This allows the actual nagios work involved with
check_send_receive_time nagios checks to be done by an unprivileged
"nagios" user rather than the "zulip" user.
2016-10-26 00:26:52 -07:00
Tim Abbott 96cf330649 puppet: ssh as the nagios user instead of zulip user.
This is a follow-up to 4f58fef54b,
touching services.cfg instead of commands.cfg.
2016-10-26 00:23:47 -07:00
Tim Abbott a350d43683 puppet: Add recovery.conf configuration to postgres_slave.pp.
This file is needed to run a valid postgres slave; it's not clear why
this wasn't installed in the original zulip.com configuration.
2016-10-26 00:22:57 -07:00
Tim Abbott c3727c9886 nagios: Remove old zulip.com trac/git/replica servers.
These are unlikely to be relevant to anyone.
2016-10-26 00:21:53 -07:00
Tim Abbott 383f39b543 nagios: Enable allow_empty_hostgroup_assignment.
This fixes the configuration being broken when we remove some of the
old zulip.com hosts that are unlikely to be of interest to anyone.
2016-10-26 00:19:21 -07:00
Tim Abbott 4f58fef54b zulip_ops: Use nagios user for all Nagios checks.
There's no reason these Nagios checks needs to run as the
semi-priviliged Zulip user.
2016-10-26 00:17:26 -07:00
Tim Abbott 32d244dbe5 puppet: Add Nagios checks for other consumers. 2016-10-26 00:11:08 -07:00
Tim Abbott f1fa4397f3 puppet: Fix package deps for zulip-ec2-configure-interfaces. 2016-10-26 00:11:08 -07:00
Tim Abbott 3448ab4c7a zulip-ec2-configure: Fix network IDs for Ubuntu Xenial. 2016-10-26 00:11:08 -07:00
Tim Abbott 080dd8c987 nagios: Ignore kthreads in check_procs tests.
Modern Linux can have a lot of kernel threads not doing anything.
Since this isn't interesting from a monitoring perpsective, we ignore
these.
2016-10-26 00:10:40 -07:00
Tim Abbott 4c9a283542 puppet: Remove configuration for old builder host.
I don't think this configuration was ever even used; it's just
clutter.
2016-10-26 00:01:52 -07:00
Tim Abbott f9ad75f58e puppet: Remove configuration for old zulip.com bots host.
This configuration didn't do anything anyway and just clutters the
repo.
2016-10-26 00:01:29 -07:00
Tim Abbott 9d4f3f1e1b puppet: Replace zulip_ops postgres configs with postgres_appdb_tuned.
There's no longer a reason to have copies of forked postgres
configuration files in our repository, since some time ago we merged
the features of these configuration files into the main
postgres_appdb_tuned.pp.
2016-10-25 23:58:53 -07:00
Tim Abbott 105ea972f6 puppet: Remove now-unncessary kernel.shm sysctl values.
With modern Linux and postgres, these settings are not required.
2016-10-25 23:58:33 -07:00
Tim Abbott 2227e77cce puppet: Remove Dropbox usernames from Nagios config. 2016-10-25 23:55:42 -07:00
Tim Abbott 8584c05d80 zulip_ops: Remove unnecessary loadbalancer stanzas. 2016-10-25 23:52:37 -07:00
Tim Abbott 624ee3989f puppet: Remove old Dropbox certificates. 2016-10-25 23:52:30 -07:00
Tim Abbott f0bb78ba2d puppet: Fix iptables-persistent->netfilter-persistent rename. 2016-10-25 23:45:21 -07:00
Tim Abbott c4ca7ee6e1 puppet: Move Apache sites files to correct paths.
Apache now actually requires its configuration files have names ending
with .conf.
2016-10-25 23:44:28 -07:00
Tim Abbott a5a03c2e0b zulip_ops: Include zulip::apt_repository.
This replaces the old wheezy configuration.
2016-10-16 20:13:35 -07:00
Tim Abbott 8c68c6f09b zulip_ops: Remove wheezy apt repo.
Nobody uses wheezy anymore, and the configuration wasn't even
conditional on the OS version.
2016-10-16 20:13:35 -07:00
Tim Abbott 5210b0a6a4 zulip_ops: Cleanup old redis configuration.
One can now just use the improved configuration we've merged into the
main Zulip repo.
2016-10-16 20:13:35 -07:00
Tim Abbott 869f0724ce zulip_ops: Remove humbughq.com nginx configuration.
The humbughq.com name hasn't been the product's name since 2013, and
it's nice to finish clearing it out of the repository.
2016-10-16 20:13:29 -07:00
Tim Abbott 29448fb47b zulip_ops: Remove old Zulip, Inc. trac configuration.
This isn't useful to anyone.
2016-10-16 19:23:47 -07:00
Tim Abbott 771e03cfa7 zulip_ops: Remove old Zulip, Inc. mediawiki configuration.
This isn't useful to anyone.
2016-10-16 19:23:47 -07:00
Tim Abbott 36e336edc3 puppet: Rename zulip_internal to zulip_ops.
The old "zulip_internal" name was from back when Zulip, Inc. had two
distributions of Zulip, the enterprise distribution in puppet/zulip/
and the "internal" SAAS distribution in puppet/zulip_internal.  I
think the name is a bit confusing in the new fully open-source Zulip
work, so we're replacing it with "zulip_ops".  I don't think the new
name is perfect, but it's better.

In the following commits, we'll delete a bunch of pieces of Zulip,
Inc.'s infrastructure that don't exist anymore and thus are no longer
useful (e.g. the old Trac configuration), with the goal of cleaning
the repository of as much unnecessary content as possible.
2016-10-16 19:23:27 -07:00