Closed Makosso closed 4 years ago
danyill
commented
4 years ago I have a similar problem. I think in the docker-compose file it is caused by the moved away from bind mounts and towards volumes (see https://github.com/sameersbn/docker-gitlab/issues/2027). There may also be other things as I was unable to upgrade from 12.1.6 to 12.3.5 even after reverting to bind mounts (which I have yet to diagnose) but you could try that.
Makosso
commented
4 years ago @danyill thanks for the tip, however using volumes in place of binds doesn't change anything. On a fresh instance, using docker volumes I'm still getting the same error, resulting in gitlab container crashing with exit code 1:
gitlab_1 | rake aborted! gitlab_1 | storage "default" is missing a gitaly_address
danyill
commented
4 years ago I found I could migrate from 12.1.6 to 12.5.2 today without any difficulty so I suspect there was an issue in perhaps the 12.3 series.
I've retained bind mounts (so changed the docker-compose.yml only from the one available in this repo around about the tag 12.1.6) and then only changed the version of the gitlab container from 12.1.6 to 12.5.2 to achieve this).
Compared to the current docker-compose.yml, for the bind mounts, the changes are something like this:
10c10
< - /srv/docker/gitlab/redis:/var/lib/redis:Z
---
> - redis-data:/var/lib/redis:Z
16c16
< - /srv/docker/gitlab/postgresql:/var/lib/postgresql:Z
---
> - postgresql-data:/var/lib/postgresql:Z
34c33
< - /srv/docker/gitlab/gitlab:/home/git/data:Z
---
> - gitlab-data:/home/git/data:Z
> volumes:
> redis-data:
> postgresql-data:
> gitlab-data:and from:
gitlab:
restart: always
image: sameersbn/gitlab:12.1.6to:
gitlab:
restart: always
image: sameersbn/gitlab:12.5.2Thank You for the mention @danyill, will definitely try the new image once I get back from holiday
f3l1x
commented
4 years ago Hi. I've just run into the same issue.
gitlab_1 | 2020-01-31 20:08:04,817 INFO success: unicorn entered RUNNING state, process has stayed up for > than 1 seconds (startsecs)
gitlab_1 | 2020-01-31 20:08:04,817 INFO success: sidekiq entered RUNNING state, process has stayed up for > than 1 seconds (startsecs)
gitlab_1 | 2020-01-31 20:08:04,817 INFO success: nginx entered RUNNING state, process has stayed up for > than 1 seconds (startsecs)
gitlab_1 | 2020-01-31 20:08:04,817 INFO success: sshd entered RUNNING state, process has stayed up for > than 1 seconds (startsecs)
gitlab_1 | 2020-01-31 20:08:04,818 INFO success: cron entered RUNNING state, process has stayed up for > than 1 seconds (startsecs)
postgresql_1 | 2020-01-31 19:09:00.693 UTC [220] ERROR: relation "feature_gates" does not exist at character 566
postgresql_1 | 2020-01-31 19:09:00.693 UTC [220] STATEMENT: SELECT a.attname, format_type(a.atttypid, a.atttypmod),
postgresql_1 | pg_get_expr(d.adbin, d.adrelid), a.attnotnull, a.atttypid, a.atttypmod,
postgresql_1 | c.collname, col_description(a.attrelid, a.attnum) AS comment
postgresql_1 | FROM pg_attribute a
postgresql_1 | LEFT JOIN pg_attrdef d ON a.attrelid = d.adrelid AND a.attnum = d.adnum
postgresql_1 | LEFT JOIN pg_type t ON a.atttypid = t.oid
postgresql_1 | LEFT JOIN pg_collation c ON a.attcollation = c.oid AND a.attcollation <> t.typcollation
postgresql_1 | WHERE a.attrelid = '"feature_gates"'::regclass
postgresql_1 | AND a.attnum > 0 AND NOT a.attisdropped
postgresql_1 | ORDER BY a.attnum
postgresql_1 |
postgresql_1 | 2020-01-31 19:09:03.586 UTC [221] ERROR: relation "feature_gates" does not exist at character 566
postgresql_1 | 2020-01-31 19:09:03.586 UTC [221] STATEMENT: SELECT a.attname, format_type(a.atttypid, a.atttypmod),
postgresql_1 | pg_get_expr(d.adbin, d.adrelid), a.attnotnull, a.atttypid, a.atttypmod,
postgresql_1 | c.collname, col_description(a.attrelid, a.attnum) AS comment
postgresql_1 | FROM pg_attribute a
postgresql_1 | LEFT JOIN pg_attrdef d ON a.attrelid = d.adrelid AND a.attnum = d.adnum
postgresql_1 | LEFT JOIN pg_type t ON a.atttypid = t.oid
postgresql_1 | LEFT JOIN pg_collation c ON a.attcollation = c.oid AND a.attcollation <> t.typcollation
postgresql_1 | WHERE a.attrelid = '"feature_gates"'::regclass
postgresql_1 | AND a.attnum > 0 AND NOT a.attisdropped
postgresql_1 | ORDER BY a.attnum
postgresql_1 |
postgresql_1 | 2020-01-31 19:09:04.602 UTC [222] ERROR: relation "feature_gates" does not exist at character 566
postgresql_1 | 2020-01-31 19:09:04.602 UTC [222] STATEMENT: SELECT a.attname, format_type(a.atttypid, a.atttypmod),
postgresql_1 | pg_get_expr(d.adbin, d.adrelid), a.attnotnull, a.atttypid, a.atttypmod,
postgresql_1 | c.collname, col_description(a.attrelid, a.attnum) AS comment
postgresql_1 | FROM pg_attribute a
postgresql_1 | LEFT JOIN pg_attrdef d ON a.attrelid = d.adrelid AND a.attnum = d.adnum
postgresql_1 | LEFT JOIN pg_type t ON a.atttypid = t.oid
postgresql_1 | LEFT JOIN pg_collation c ON a.attcollation = c.oid AND a.attcollation <> t.typcollation
postgresql_1 | WHERE a.attrelid = '"feature_gates"'::regclass
postgresql_1 | AND a.attnum > 0 AND NOT a.attisdropped
postgresql_1 | ORDER BY a.attnum
postgresql_1 |
gitlab_1 | Database 'gitlabhq_production' already exists
postgresql_1 | 2020-01-31 19:09:13.633 UTC [226] ERROR: database "gitlabhq_production" already exists
postgresql_1 | 2020-01-31 19:09:13.633 UTC [226] STATEMENT: CREATE DATABASE "gitlabhq_production" ENCODING = 'unicode'
gitlab_1 | Migrating database...
gitlab_1 | Clearing cache...
gitlab_1 | 2020-01-31 20:11:42,427 WARN received SIGTERM indicating exit request
gitlab_1 | 2020-01-31 20:11:42,433 INFO waiting for gitaly, gitlab-workhorse, unicorn, sidekiq, nginx, sshd, cron to die
gitlab_1 | 2020-01-31 20:11:42,434 INFO stopped: sshd (exit status 0)
gitlab_1 | 2020-01-31 20:11:42,434 INFO stopped: cron (terminated by SIGTERM)I've took docker-compose.yml and changed volumes.
gitlab:
restart: always
image: sameersbn/gitlab:12.7.0
depends_on:
- redis
- postgresql
ports:
- "22:22"
volumes:
- ./data/gitlab:/home/git/data:ZWell, it works eventually even when error is showed. Weird. :-)
stale[bot]
commented
4 years ago This issue has been automatically marked as stale because it has not had any activity for the last 60 days. It will be closed if no further activity occurs during the next 7 days. Thank you for your contributions.
Hi! I've been using this image for quite some time, however it refuses to upgrade, or even launch with no data from image version 12.3 and up. The error I'm getting is:
The same thing happens (rake aborted/storage default is missing a gitaly_address) both when upgrading the image on previously working instance (tried upgrading from 11.11.0 and 12.1.6) and on a fresh deploy, with no data. My current instance based on 12.1.6 image works with no problem whatsoever.
os: ubuntu 18.04 4.15.0-34-generic docker version: 18.06.1-ce, build e68fc7a docker-compose version: 1.17.1, build unknown
Any clue as to what's the cause of this behavior? Is there a particular thing I need to fix on my end in order to upgrade to 12.3.x and further?