Commit Graph

1826 Commits

Author SHA1 Message Date
Jessica McKellar
61d660f9f3 [manual] digest: move cron job from staging to all app frontends.
Manual instructions:

This commit requires a puppet apply after deployment on both staging
and prod.

(imported from commit 2d10e33c6db2f5e9cc1204cdd5f2c91833da2a8e)
2013-12-20 12:50:23 -05:00
Tim Abbott
bdcc2e5c52 nagios: Set max_check_attempts to 3 for batched queue processors.
(imported from commit ec0ac86726cd6ff3d0fdfcfcb161d3329fca02ac)
2013-12-19 17:31:41 -05:00
Tim Abbott
b2d01e2da0 [manual] restart-server: Minimize downtime for message sender worker.
The manual step here is that we need to do the `puppet apply` before
pushing this commit, or `restart-server` will crash.

Previously we shut down everything in one group, which performed
poorly with supervisor's bad performance on restarting many daemons at
once.  Now we shut down the unimportant stuff, then the important
stuff, bring back the important stuff, and then bring back the
unimportant stuff.

This new model has a little over 5s of downtime for the core
user-facing daemons -- which is still far more than would be ideal,
but a lot less than the 13s or so that we had before.

Here's some logs with the current setup for the tornado/django downtime:
2013-12-19 20:16:51,995 restart-server: Stopping daemons
2013-12-19 20:16:53,461 restart-server: Starting daemons
2013-12-19 20:16:57,146 restart-server: Starting workers

Compare with the behavior on master today:
2013-12-19 20:21:45,281 restart-server: Stopping daemons
2013-12-19 20:21:49,225 restart-server: Starting daemons
2013-12-19 20:21:58,463 restart-server: Done!

(imported from commit b2c1ba77f3dc989551d0939779208465a8410435)
2013-12-19 17:21:23 -05:00
Luke Faraone
a5775d94ef Install uploads.types in puppet
We also move uploads.types to zulip-include-frontend since its only
needed on the frontends.

(imported from commit cfdf15c0c537f7ea4c239b0f882aeaa561929777)
2013-12-18 16:25:10 -05:00
Leo Franchi
9c82e869c2 [manual] Release OS X desktop app 0.4.2
This reqires a puppet apply as well as a manual move of the installed
files and symlink switch. Leo will do it when it hits master.

(imported from commit e58e52087ad38f1cb8e0e606b82266a93cf91e53)
2013-12-18 16:14:51 -05:00
Jessica McKellar
5e217a1079 Use correct time zone in digest email cron job.
(imported from commit fd470af4b44ffb9696ff3a97372aaf2524a4806b)
2013-12-18 14:31:03 -05:00
Tim Abbott
d62ca820db puppet: Fix permissions on /etc/cron.d/log2zulip.
(imported from commit 33ee5ae97b09b3925849940262ecd0bcbce38a3f)
2013-12-17 16:22:14 -05:00
Tim Abbott
ae6c17a87d puppet: Stop using /var/log/nginx/zulip.*.log.
It's confusing to have our log data on different files on different
systems (e.g. loadbalancer vs. app).

(imported from commit be701072ee05e2659f146b226a39f33cb4707180)
2013-12-17 16:22:08 -05:00
Tim Abbott
6ccf19bed6 Run log2zulip on load balancers too.
(imported from commit 74c8be20d2d03aa524f05b7681febe9a9be9cdff)
2013-12-17 13:46:00 -05:00
Tim Abbott
8dcf7d4cc3 [puppet] Add log2zulip tool for sending log files to Zulip.
This tool is a little crude; it runs out of a cron job and will
forward to staging a notice about any new lines in the declared log
files, truncating if there are more than 10 lines.

(imported from commit 6748ddff1def0907b061dc278a3a848bd2e933f1)
2013-12-17 11:02:55 -05:00
Jessica McKellar
8bb1caec8f [manual] digest: add the cron job that will trigger digest emails.
Manual deployment instructions:

On staging, do a puppet apply.

No action needs to be taken for the prod deploy.

(imported from commit 0f6e5ab22aaeacfcc69d57de12f2bb6fac6f0635)
2013-12-17 10:47:16 -05:00
Tim Abbott
b6acbe040c Fix missing nginx service notifications on configuration changes.
(imported from commit 0bfce276bab3704e508f6c8a58c9434e9fc224cd)
2013-12-16 13:44:50 -05:00
Tim Abbott
c872866289 puppet: Fix nginx upstreams for staging.
(imported from commit eb1e6e3b2d35533af4a24015a91201e2414f8e28)
2013-12-16 11:32:05 -05:00
Tim Abbott
f8fe9d1dd4 Fix check_worker_memory process list computation.
(imported from commit 9ac58b894ecfd84da6ac8509c0dc2ceb60eedfce)
2013-12-16 10:09:59 -05:00
Luke Faraone
1370c014a5 Clean up logging and documentation in ec2 interfaces script
(imported from commit e55247931cdeb61563f2348ca09f3d7b9fc85f0c)
2013-12-13 18:07:08 -05:00
Luke Faraone
104c2a06ae Set iptables rules for each IP, not just each interface
(imported from commit c24d2123489dc384bf50e379d245807af3488ebf)
2013-12-13 18:07:08 -05:00
Kevin Mehall
f929e51776 puppet: Make Camo Nagios check waste less bandwidth
Use http://www.google.com/favicon.ico instead of a 1.7MB animated gif from
imgur.

(imported from commit 94993af35bf87b0f22e6e743a9ba1cc1c5c9a78f)
2013-12-13 17:27:01 -05:00
Tim Abbott
950e4c800b puppet: Declare upstreams properly in app nginx config.
(imported from commit 859eeed0d5b92c1b5b2b0764aba06aebcde8e2e2)
2013-12-12 16:48:52 -05:00
Tim Abbott
ae4d214c49 Fix longpolling treatment for api.zulip.com/v1/events.
(imported from commit 78029972938ad7c9aa862330e38965b4b032c935)
2013-12-12 16:03:45 -05:00
Tim Abbott
73f04b21e9 Add zulip.customer29.invalid host.
(imported from commit ea3e7bb465c920b8ec21b7471cd261868f5059e7)
2013-12-12 16:03:45 -05:00
Tim Abbott
c21e85e569 Remove staging.humbughq.com loadbalancer config.
The DNS has been disabled for some time.

(imported from commit e054c0fb0b37077d8303eab4d4ffec6ff53e8990)
2013-12-12 16:03:45 -05:00
Kevin Mehall
662edc2558 [manual] Backend support for Android GCM push notifications
This adds a dependency on gcmclient:
http://gcm-client.readthedocs.org/en/latest/gcmclient.html

pip install gcm-client

or

apt-get install python-gcm-client

(imported from commit 9f1fbf1f793e4a27baed85c6f1aa7a7b03106a10)
2013-12-11 15:37:48 -05:00
Luke Faraone
b0a0853bd2 Specify full fingerprint rather than short key ID
(imported from commit fc4e9d51c440000e469f8e3882739215a3bcb022)
2013-12-11 10:54:30 -05:00
Luke Faraone
510b3349a7 Switch to downloading keys via SSL in puppet
(imported from commit 05d2b0626338f09370614e916050cfcee7f14829)
2013-12-11 10:54:30 -05:00
Luke Faraone
1b5c1ac021 Update style of client strings.
(imported from commit 1516461cf53b2715de68e01f16bb8a8cc33c48ad)
2013-12-09 11:47:52 -05:00
Leo Franchi
e39cc5324b [puppet] Aggregate narrow timing stats
(imported from commit 4eff25635a3cb7687e995ad1127cff68da51329a)
2013-12-07 10:44:54 -05:00
Leo Franchi
f70878e6c5 Fix aggregation rules for endtoend time
(imported from commit 29165b09e2d8904ee502cc04610a951d87ef896f)
2013-12-07 10:44:54 -05:00
Tim Abbott
abeb29c226 Fix incorrect proxy_pass location for staging longpolling.
(imported from commit a4ac2c5c3416a8d8f748237411df6235f237e893)
2013-12-07 08:02:55 -05:00
Tim Abbott
09a61e8128 nginx: Enable keepalive for communication between lbs and frontends.
(imported from commit a7c8d9dfefbb6e5d01c8050688d831787b31bbd4)
2013-12-07 07:41:45 -05:00
Tim Abbott
a337638f7f nginx config: Enable some popular performance-improving features.
(imported from commit 50193d071d091cc864867c9f7d5c9c9fb74b9d92)
2013-12-07 07:41:45 -05:00
Tim Abbott
6663247e22 Set nginx worker_processes automatically based on CPU count.
(imported from commit e48143c1410439d0574bc78bfd64c22a3063d558)
2013-12-07 07:41:45 -05:00
Tim Abbott
1843262672 puppet: Mark all Nagios plugins as executable.
They were being installed as executable anyway, but this will make
running them manually a bit easier.

(imported from commit a1181d2c90770af5aa44b0f65a47a460efdcf2d7)
2013-12-05 15:25:25 -05:00
Tim Abbott
64807c0628 nginx: Ensure zulip-include files are distributed to the right systems.
There were a few recently introduced bugs, and this also cuts down on
our having to review diffs that don't actually affect the relevant
server when doing updates.

(imported from commit 43f3cff9a414bc1632f45a8222012846353e8501)
2013-12-05 15:25:25 -05:00
Tim Abbott
676e9d90ff nginx: Get rid of trailing / in loadbalancer proxy_pass directives.
The trailing "/" actually means "replace the location with /", which
is either useless or actively harmful, depending on the location.

(imported from commit 58b9c4c9e55e3a162ffce49c954bc2182ec57dde)
2013-12-05 15:25:25 -05:00
Tim Abbott
cc00ed6d7e nginx: Clean up now-empty 'loadbalancer' include file.
(imported from commit d13b5d91f6b85ba3e0bef7728985d0eba1cae084)
2013-12-05 15:25:25 -05:00
Tim Abbott
afaff0c2cf nginx: Set X-Forwarded-For in common proxy configuration.
Previously we sometimes set it to $proxy_add_x_forwarded_for and other
times to $remote_addr, but according to

http://wiki.nginx.org/HttpProxyModule#.24proxy_add_x_forwarded_for

$proxy_add_x_forwarded_for handles this for us -- it will be
$remote_addr if there was no X-Forwarded-For header anyway.

(imported from commit 67dc52250e3e7751b1bf375d1a71d0272475435c)
2013-12-05 15:25:25 -05:00
Tim Abbott
afe167ea58 nginx: Use the longpolling proxy configuration on load balancers.
(imported from commit f590e6b1eec2856b5128e310797f8ba58846417a)
2013-12-05 15:25:25 -05:00
Tim Abbott
21a69f2188 nginx: Move common longpolling proxy configuration into include file.
(imported from commit 4ace82824c32cec8c6da8a1a6b8a527dae105a89)
2013-12-05 15:25:24 -05:00
Tim Abbott
9e24558092 nginx: Move common proxy configuration into an include file.
(imported from commit 2ee5afc74fe146f8ee98f18f846342351c61c7f0)
2013-12-05 15:25:24 -05:00
Tim Abbott
3760609f3f Enable /sockjs handling on api.zulip.com (not used yet).
(imported from commit c2581e3243b2129c980fd3dd318eb3d99f3eb593)
2013-12-05 15:25:24 -05:00
Tim Abbott
953c3578dc external-sso.conf: Fix missing proxy_set_header.
(imported from commit 64bcb06cf65f15908ee74d637ab3868916b1dfd7)
2013-12-05 15:25:24 -05:00
Tim Abbott
79910fa2b3 Disable proxy_next_upstream for sockjs in remaining proxy_pass lines.
(imported from commit f14c7962253b34040ed9ab077a58c8b200df5d9d)
2013-12-05 15:25:24 -05:00
Tim Abbott
e5be713103 Clean up EXTERNAL_API_HOST usage and defaults.
We now have 2 variablse:
EXTERNAL_API_PATH: e.g. staging.zulip.com/api
EXTERNAL_API_URI: e.g. https://staging.zulip.com/api

The former is primarily needed for certain integrations.

(imported from commit 3878b99a4d835c5fcc2a2c6001bc7eeeaf4c9363)
2013-12-04 15:10:54 -05:00
Tim Abbott
b8a151ca4e Revert "[puppet] Add cron job to restart our workers daily."
This reverts commit 0b0180b0751f6c618d877b9c9ffc2b8287254e4d.

(imported from commit a81c552100345d369ffcaf69f28a86dea0893128)
2013-12-04 10:27:45 -05:00
Tim Abbott
f7ac58bab5 Revert "Add user_activity_test worker that does nothing."
Now that we've debugged the memory leak, I don't think we need this
anymore.

This reverts commit 1bdc7ee2f72bdebb1cdc94601247834a434614d6.

Conflicts:

	puppet/zulip/files/cron.d/rabbitmq-numconsumers
	puppet/zulip/files/supervisor/conf.d/zulip.conf

(imported from commit ff87f2aebcbc71013fa7a05aedb24e2dcad82ae6)
2013-12-04 10:27:45 -05:00
Tim Abbott
606d8a4f9b Add Nagios check for queue worker memory usage.
This is detect future memory leaks.

(imported from commit 75fd4c2ad41ea71e87a53fb33e2106c5773909d5)
2013-12-04 10:27:44 -05:00
Tim Abbott
850eae3e8e puppet: Disable proxy_next_upstream feature in nginx config.
(imported from commit 84cad76701f9ee40fa9601ae06b3f804948b96d4)
2013-12-03 15:20:45 -05:00
Tim Abbott
5007d4d87a [puppet] Update set_real_ip_from to use lb0's internal IP address.
This is something we forgot to do in the VPC migration, so our IPs
have all been the lb0 IP in our logs :(.

(imported from commit 9d3fc69cf72a84f7bd7c54e50fb1e776a67d971f)
2013-12-03 14:29:34 -05:00
Luke Faraone
dc5c410257 Remove django-jstemplate, unused per trac #1973
(imported from commit 97b2e75766e3576f17b7dab0f1a4a00c34a5c2e5)
2013-12-03 11:36:18 -05:00
Leo Franchi
42e23dc82e [manual] Release desktop app 0.4.1 for OS X
This requires a puppet apply on prod0, and an update of the
Zulip-latest.dmg and Humbug-latest.dmg symlinks in
/src/www/dist/apps/mac and /srv/www/dist/apps/sso/mac

(imported from commit e83170a19ac2de6458a0fd43140068fab4135483)
2013-12-02 15:24:32 -05:00