~hloeung/ubuntu-repository-cache/install-w3m-for-apache2ctl : changes

~hloeung/ubuntu-repository-cache/install-w3m-for-apache2ctl » Changes from revision 342

From Revision 321 to 302

Rev	Summary	Authors	Date
321	Include syncing dist-upgrade-all metadata - LP:1765934 Include syncing dist-upgrade-all metadata - LP:1765934 Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/sync-dist-upgrade-all-metadata/+merge/398938 Reviewed-by: Colin Misare <colin.misare@canonical.com>	"Haw Loeung"	3 years ago
320	We're already logging, so may as well log to the right place... We're already logging, so may as well log to the right place - LP:1917385 Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/fix-squid-access-log-location/+merge/399035 Reviewed-by: Paul Collins <paul.collins@canonical.com>	"Haw Loeung"	3 years ago
319	Add socket timeout for sending metrics to influx Add socket timeout for sending metrics to influx Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/add-send-to-influx-socket-timeout/+merge/398931 Reviewed-by: Paul Collins <paul.collins@canonical.com>	"Haw Loeung"	3 years ago
318	Add apache2 cgi script timeout Add apache2 cgi script timeout Otherwise, value of `Timeout` is used and that defaults to 60secs which is too long. We have two check points (metadata and pool) with a timeout in the health-check cgi script to 10secs for each so 20secs should be enough for both. Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/apache2-cgi-script-timeout/+merge/398920 Reviewed-by: Benjamin Allot <benjamin.allot@canonical.com>	"Haw Loeung"	3 years ago
317	Bump metadata sync timeout Bump metadata sync timeout As seem in AWS us-east-1, it can take over an hour to sync the metadata (which is now 26GiB) from another unit in the same region. Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/bump-metadata-sync-timeout/+merge/398848 Reviewed-by: Max Horlanchuk <max.horlanchuk@canonical.com>	"Haw Loeung"	3 years ago
316	Only reload/graceful apache2 when required - LP:1916084 Only reload/graceful apache2 when required - LP:1916084 Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/only-reload-apache2-when-needed/+merge/398325 Reviewed-by: Joel Sing <joel.sing@canonical.com>	"Haw Loeung"	3 years ago
315	unittests now uses hookenv.Config() so ensure .juju-persiste... unittests now uses hookenv.Config() so ensure .juju-persistent-config isn't included or is properly cleaned Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/cleanup-for-charm-build/+merge/398499 Reviewed-by: Colin Misare <colin.misare@canonical.com>	"Haw Loeung"	3 years ago
314	Don't serve health-check files if cgid module isn't load Don't serve health-check files if cgid module isn't load Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/missing-apache2-cgi-mod-2/+merge/398492 Reviewed-by: Benjamin Allot <benjamin.allot@canonical.com>	"Haw Loeung"	3 years ago
313	Enable cgi module on install otherwise health check succeeds... Enable cgi module on install otherwise health check succeeds with output from the script; also we want cgid for multi-threaded apache2 Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/missing-apache2-cgi-mod/+merge/398421 Reviewed-by: Paul Collins <paul.collins@canonical.com>	"Haw Loeung"	3 years ago
312	Don't call apache.start() in service.unpause() Don't call apache.start() in service.unpause() service.unpause() calls render_configs() which calls apache.render_configs(). Within that, it will call apache.start() so it is not needed here. It also handles config changes requiring graceful reload better. Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/only-reload-apache2-when-needed-4/+merge/398419 Reviewed-by: Paul Collins <paul.collins@canonical.com>	"Haw Loeung"	3 years ago
311	Only reload/graceful apache2 when required - LP:1916084 Only reload/graceful apache2 when required - LP:1916084 Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/only-reload-apache2-when-needed-3/+merge/398328 Reviewed-by: Joel Sing <joel.sing@canonical.com>	"Haw Loeung"	3 years ago
310	Only reload/graceful apache2 when required - LP:1916084 Only reload/graceful apache2 when required - LP:1916084 Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/only-reload-apache2-when-needed-2/+merge/398326 Reviewed-by: Barry Price <barry.price@canonical.com>	"Haw Loeung"	3 years ago
309	Handle when newly provisioned unit Handle when newly provisioned unit Newly provisioned units set the charm's failover feature. This sets up proxying of all requests to the last u-r-c unit. This means that the health-check will succeed as it's checking on the previous unit. We want to fix this by not proxying the health-check endpoint URL (/_health-check/health-check.py). We also want to set the health-check's 'disabled for maintenance' flag when the charm's failover feature is activated. Finally, update the health-check to ensure ubuntu_active is present. It's not present on initial metadata sync. Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/health-check-handle-initial-sync/+merge/398100 Reviewed-by: Nick Moffitt <nick.moffitt@canonical.com> Reviewed-by: Tom Haddon <tom.haddon@canonical.com> Reviewed-by: Laurent Sesques <laurent.sesques@canonical.com>	"Haw Loeung"	3 years ago
308	Fix issue with initial metadata sync failing - LP:1803453 Fix issue with initial metadata sync failing - LP:1803453 Initial sync fails with SSH Host key verification failed. This is due to the copied host key from the leader unit not including itself[1]. Instead, we generate a list of hosts from the relation and gather keys using the same method, unison.write_known_hosts(). [1]https://github.com/juju/charm-helpers/blob/master/charmhelpers/contrib/unison/__init__.py#L204 Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/ssh-known-hosts-on-join/+merge/398095 Reviewed-by: Paul Collins <paul.collins@canonical.com>	"Haw Loeung"	3 years ago
307	503, as in HTTP/503 for disabled units for maintenance 503, as in HTTP/503 for disabled units for maintenance	Haw Loeung	3 years ago
306	Offset DISABLED (or units marked down for maintenance) metri... Offset DISABLED (or units marked down for maintenance) metric from health check endpoint so it's clearly visible on grafana graphs Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/health-check-endpoint-2/+merge/397885 Reviewed-by: Barry Price <barry.price@canonical.com>	"Haw Loeung"	3 years ago
305	Ensure logging directory exists before apache.render_configs... Ensure logging directory exists before apache.render_configs() which relies on it Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/ensure-logging-dir-present-before-render-configs/+merge/397879 Reviewed-by: Paul Collins <paul.collins@canonical.com>	"Haw Loeung"	3 years ago
304	Enable health-check endpoint by default Enable health-check endpoint by default	Haw Loeung	3 years ago
303	New health check endpoint for load balancers to use New health check endpoint for load balancers to use Load balancers can either monitor metadata (e.g. ubuntu/dists/*) or pool files, which are proxied to the real archive servers. Unfortunately, most can only check one thing so if the other is having issues, it is not taken offline and still being used. The new health check endpoint checks a few things and reports back OK if all are healthy and fine. This means load balancers can just hit this one endpoint and will take it offline if any of the components are experiencing issues. Reviewed-on: https://code.launchpad.net/~hloeung/ubuntu-repository-cache/health-check-endpoint/+merge/397547 Reviewed-by: Barry Price <barry.price@canonical.com> Reviewed-by: Benjamin Allot <benjamin.allot@canonical.com>	"Haw Loeung"	3 years ago
302	Make it clear this is the charm source version/commit Make it clear this is the charm source version/commit	Haw Loeung	3 years ago