Commit Graph

286 Commits

Author SHA1 Message Date
Alex Vandiver a9183d2208 grafana: Enable auto-sign-up.
This avoids the need to explicitly create new users in Grafana, by
simply trusting Teleport.
2022-07-19 17:52:17 -07:00
Alex Vandiver 9bd88a93e2 puppet: Tell needrestart to not default to restarting core services.
The `needrestart` tool added in 22.04 is useful in terms of listing
which services may need to be restarted to pick up updated libraries.
However, it prompts about the current state of services needing
restart for *every* subsequent `apt-get upgrade`, and defaulting core
services to restarting requires carefully manually excluding them
every time, at risk of causing an unscheduled outage.

Build a list of default-off services based on the list in
unattended-upgrades.
2022-07-19 17:51:18 -07:00
Alex Vandiver 7ae3708c02 teleport: Add explicit WebAuthn config, not just U2F.
WebAuthn is the default, replacing U2F, in Teleport 10 and above[1].
While Teleport can derive a WebAuthn configuration from a U2F
configuration[2], it's useful to be explicit.

[1]: https://goteleport.com/docs/access-controls/guides/webauthn/
[2]: https://goteleport.com/docs/access-controls/guides/webauthn/#u2f
2022-07-18 11:41:00 -07:00
Alex Vandiver 120de1dca9 zephyr: Write out unix timestamp in check, as check_cron_file expects.
A follow-up fix to 8bc26aab08.
2022-06-30 11:12:26 -07:00
Alex Vandiver 8577adcf2e cron: Remove unused STATE_FILE environment variable. 2022-06-22 12:07:38 -07:00
Alex Vandiver 8bc26aab08 nagios: Switch check_user_zephyr_mirror_liveness to run via cron.
This check loads Django, and as such must be run as the zulip user.
Repeat the same pattern used elsewhere in nagios, of writing a state
file, which is read by `check_cron_file`.
2022-06-22 12:07:38 -07:00
Alex Vandiver 41deef40cf nagios: Switch to generic check_cron_file for queues and consumers.
These share a common root; 91da4bd59b duplicated the code, but
didn't move the existing uses to the new utility.
2022-06-22 12:07:38 -07:00
Alex Vandiver 8fbde9b8c5 nagios: Only run check_fts_update_log on one PostgreSQL host.
The data is the same in the table in all replicas -- there is no need
to alert on all of them.
2022-06-22 12:07:38 -07:00
Alex Vandiver 499284d2fd nagios: Split postgresql into primary and replica.
Replication checks should only run on primary and replicas, not
standalone hosts; while `autovac_freeze` currently only runs on
primary hosts, it functions identically on replicas, and is fine to
run there.

Make `autovac_freeze` run on all `postgresql` hosts, and make
standalone hosts no longer `postgres_primary`, so they do not fail the
replication tests.
2022-06-22 12:07:38 -07:00
Alex Vandiver 38e435347b nagios: Add missing queue consumer checks. 2022-06-22 12:07:38 -07:00
Alex Vandiver e01a4242aa nagios: Sort queue consumer checks. 2022-06-22 12:07:38 -07:00
Alex Vandiver 2c90c7a010 nagios: Switch `check_remote_arg_string` queue checks to consumer checks.
These style of checks just look for matching process names using
`check_remote_arg_string`, which dates to 8edbd64bb8.  These were
added because the original two (`missedmessage_emails` and
`slow_queries`) did not create consumers, instead polling for events.

Switch these to checking the queue consumer counts that the
`check-rabbitmq-consumers` check is already writing out.  Since the
`missedmessage_emails` was _already_ checked via the consumer check, a
duplicate is not added.
2022-06-22 12:07:38 -07:00
Alex Vandiver f48d543d9b nagios: Make and use a "rabbitmq-consumer-service" template service. 2022-06-22 12:07:38 -07:00
Alex Vandiver 775a084d0f nagios: Add a catchall "other" set. 2022-06-22 12:07:38 -07:00
Alex Vandiver 83c82c8e15 nagios: Adjust load alerting by hostgroup.
Even the `pageable_servers` group did not page for high load -- in
part because what was "high" depends on the servers.  Set slightly
better limits based on server role.
2022-06-22 12:07:38 -07:00
Alex Vandiver 2a14aa5180 nagios: Add a `fullstack` hostgroup.
This will be used to apply checks only to czo.
2022-06-22 12:07:38 -07:00
Alex Vandiver b5ecfc327f nagios: Remove unnecessary `web` hostgroup.
This had identical membership to `frontends`.
2022-06-22 12:07:38 -07:00
Alex Vandiver d9d0014fb4 nagios: Rename `zmirror_main` into `zmirror` hostgroup.
`zmirror` itself was `zmirror_main` + `zmirrorp` but was unused; we
consistently just use the term `zmirror` for the non-personals server,
so use it as the hostgroup name.
2022-06-22 12:07:38 -07:00
Alex Vandiver 70c36985b4 nagios: Remove frontends from redis group.
The Redis nagios checks themselves are done against `redis` +
`frontends` groups, so there is no need to misleadingly place
`frontends` in the `redis` hostgroup.
2022-06-22 12:07:38 -07:00
Alex Vandiver 4c17f2bccc nagios: The frontends hostgroup now includes prod and staging frontends.
This lets the config file remove some repetition.
2022-06-22 12:07:38 -07:00
Alex Vandiver 1e81775fa0 nagios: Drop unhelpful hostgroup comment. 2022-06-22 12:07:38 -07:00
Alex Vandiver 7b584401ac nagios: Reformat hostgroups. 2022-06-22 12:07:38 -07:00
Alex Vandiver 93bcb86345 nagios: Reorder service checks. 2022-06-22 12:07:38 -07:00
Alex Vandiver eaaa2fbff8 nagios: Use canonical "hostgroup_name" consistently. 2022-06-22 12:07:38 -07:00
Alex Vandiver e8996b53a5 nagios: Remove unused has_swap hostgroup. 2022-06-22 12:07:38 -07:00
Alex Vandiver bc4f4b4862 nagios: Make the pageable/not/flaky tri-state clearer. 2022-06-22 12:07:38 -07:00
Alex Vandiver c74f195fba nagios: Split AWS and non-AWS hosts, for ntp checks.
The non-AWS hosts cannot use the AWS ntp server for their check.
2022-06-22 12:07:38 -07:00
Alex Vandiver 872efdee58 nagios: Fold single- and multitornado_frontends back into frontends.
5abf4dee92 made this distinction, then multitornado_frontends was
never used; the singletornado_frontends alerting worked even for the
multiple-Tornado instances.

Remove the useless and misleading distinction.
2022-06-22 12:07:38 -07:00
Alex Vandiver 3741c1c034 puppet: Switch to checking time against the AWS timeserver.
Since this is what chrony is sync'ing to, it lessens the chance of
spurious firings of this alert.

See https://aws.amazon.com/blogs/aws/keeping-time-with-amazon-time-sync-service/
2022-05-31 22:57:32 -07:00
Anders Kaseorg e9ba9b0e0d zulip-ec2-configure-interfaces: Remove.
Our current EC2 systems don’t have an interface named ‘eth0’, and if
they did, this script would do nothing but crash with ImportError
because we have never installed boto.utils for Python 3.

(The message of commit 2a4d851a7c made
an effort to document for future researchers why this script should
not have been blindly converted to Python 3.  However, commit
2dc6d09c2a (#14278) was evidently
unresearched and untested.)

Signed-off-by: Anders Kaseorg <anders@zulip.com>
2022-05-03 02:25:59 -07:00
Alex Vandiver 35db1ee435 puppet: Only include "app_service" section if there are apps.
This works around gravitational/teleport#12256, but also produces config
files that are slightly cleaner.
2022-04-26 16:36:13 -07:00
Alex Vandiver f6d27562fa puppet: Configure chrony to use AWS-local NTP sources.
This prevents hosts from spewing traffic to random hosts across the
Internet.
2022-03-25 17:07:53 -07:00
Alex Vandiver 1bd5723cd2 puppet: Add a prometheus monitor for tornado processes. 2022-03-20 16:12:11 -07:00
Anders Kaseorg f6a701090c setup-apt-repos: Don’t install lsb_release.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
2022-02-14 16:38:53 -08:00
Alex Vandiver b3f07cc98d puppet: Replace debathena zephyr package with equivalent puppet file. 2022-01-18 14:13:28 -08:00
Alex Vandiver a6d7539571 puppet: Replace debathena krb5 package with equivalent puppet file. 2022-01-18 14:13:28 -08:00
Alex Vandiver fc1adef28a puppet: Fix server_name of internal staging server. 2022-01-18 12:36:56 -08:00
Alex Vandiver 7e630b81f8 puppet: Switch to using snakeoil certs for staging.
This parallels ba3b88c81b, but for the
staging host.
2022-01-18 12:36:56 -08:00
Alex Vandiver 4d7e6b26df puppet: Provide more attributes to teleport on ssh nodes. 2022-01-12 14:15:45 -08:00
Alex Vandiver 339e70671c puppet: Switch Grafana to Grafana 8 Unified Alerting. 2022-01-11 14:27:11 -08:00
Alex Vandiver 6a7eecee9a puppet: Increase load paging thresholds. 2022-01-11 09:38:31 -08:00
Alex Vandiver 3c95ad82c6 puppet: Upgrade to nagios4.
This updates the puppeted nagios configuration file for the Nagios4
defaults.
2022-01-11 09:38:31 -08:00
Alex Vandiver bb5a2c8138 puppet: Move prometheus to external_dep. 2021-12-29 16:35:15 -08:00
Alex Vandiver 2d6c096904 puppet: Move node_exporter to external_dep. 2021-12-29 16:35:15 -08:00
Alex Vandiver 053682964e puppet: Only fetch from running hosts in Grafana ec2 discovery. 2021-12-09 08:12:03 -08:00
Alex Vandiver 291f688678 puppet: Use zulip::external_dep for grafana, template config.
Templating the config ensures that the service is restarted when it is
upgraded.
2021-12-08 20:58:10 -08:00
Alex Vandiver fb2d05f9e3 puppet: Remove unused 'builder' files.
These are leftover detritus from the "builder" host, which was removed
in 4c9a283542.
2021-12-06 10:21:50 -08:00
Alex Vandiver b982222e03 camo: Replace with go-camo implementation.
The upstream of the `camo` repository[1] has been unmaintained for
several years, and is now archived by the owner.  Additionally, it has
a number of limitations:
 - It is installed as a sysinit service, which does not run under
   Docker
 - It does not prevent access to internal IPs, like 127.0.0.1
 - It does not respect standard `HTTP_proxy` environment variables,
   making it unable to use Smokescreen to prevent the prior flaw
 - It occasionally just crashes, and thus must have a cron job to
   restart it.

Swap camo out for the drop-in replacement go-camo[2], which has the
same external API, requiring not changes to Django code, but is more
maintained.  Additionally, it resolves all of the above complaints.

go-camo is not configured to use Smokescreen as a proxy, because its
own private-IP filtering prevents using a proxy which lies within that
IP space.  It is also unclear if the addition of Smokescreen would
provide any additional protection over the existing IP address
restrictions in go-camo.

go-camo has a subset of the security headers that our nginx reverse
proxy sets, and which camo set; provide the missing headers with `-H`
to ensure that go-camo, if exposed from behind some other non-nginx
load-balancer, still provides the necessary security headers.

Fixes #18351 by moving to supervisor.
Fixes zulip/docker-zulip#298 also by moving to supervisor.

[1] https://github.com/atmos/camo
[2] https://github.com/cactus/go-camo
2021-11-19 15:58:26 -08:00
Alex Vandiver 27881babab puppet: Increase prometheus storage, from the default 15d. 2021-08-24 23:40:43 -07:00
Alex Vandiver 5857dcd9b4 puppet: Configure ip6tables in parallel to ipv4.
Previously, IPv6 firewalls were left at the default all-open.

Configure IPv6 equivalently to IPv4.
2021-08-24 16:05:46 -07:00