- Issues to be aware of when upgrading from 15.11
- 16.4.0
- 16.3.0
- 16.2.0
- 16.1.0
- 16.0.0
- Long-running user type data change
- Undefined column error upgrading to 16.2 or later
GitLab 16 changes
This page contains upgrade information for minor and patch versions of GitLab 16. Ensure you review these instructions for:
- Your installation type.
- All versions between your current version and your target version.
For more information about upgrading GitLab Helm Chart, see the release notes for 7.0.
Issues to be aware of when upgrading from 15.11
- Some GitLab installations must upgrade to GitLab 16.0 before upgrading to any other version. For more information, see Long-running user type data change.
- Other installations can skip 16.0, 16.1, and 16.2 as the first required stop on the upgrade path is 16.3. Review the notes for those intermediate versions.
- If your GitLab instance upgraded first to 15.11.0, 15.11.1, or 15.11.2 the database schema is incorrect. Recommended: perform the workaround before upgrading to 16.x. See the details and workaround.
-
Linux package installations must change Gitaly and Praefect configuration structure before upgrading to GitLab 16. To avoid data loss reconfigure Praefect first, and as part of the new configuration, disable metadata verification. Read more:
16.4.0
-
Updating a group path received a bug fix that uses a database index introduced in 16.3.
If you upgrade to 16.4 from a version lower than 16.3, you must execute
ANALYZE packages_packages;
in the database before you use it. -
You might encounter the following error while upgrading to GitLab 16.4 or later:
main: == 20230830084959 ValidatePushRulesConstraints: migrating ===================== main: -- execute("SET statement_timeout TO 0") main: -> 0.0002s main: -- execute("ALTER TABLE push_rules VALIDATE CONSTRAINT force_push_regex_size_constraint;") main: -> 0.0004s main: -- execute("RESET statement_timeout") main: -> 0.0003s main: -- execute("ALTER TABLE push_rules VALIDATE CONSTRAINT delete_branch_regex_size_constraint;") rails aborted! StandardError: An error has occurred, all later migrations canceled: PG::CheckViolation: ERROR: check constraint "delete_branch_regex_size_constraint" of relation "push_rules" is violated by some row
These constraints might return an error:
author_email_regex_size_constraint
branch_name_regex_size_constraint
commit_message_negative_regex_size_constraint
commit_message_regex_size_constraint
delete_branch_regex_size_constraint
file_name_regex_size_constraint
force_push_regex_size_constraint
To fix the error, find the records in the
push_rules
table that exceed the 511 character limit.;; replace `delete_branch_regex` with a name of the field used in constraint SELECT id FROM push_rules WHERE LENGTH(delete_branch_regex) > 511;
Reduce the value length of the regex field for affected push rules records, then retry the migration.
Self-compiled installations
- A new method of configuring paths for the GitLab secret and custom hooks is preferred in GitLab 16.4 and later:
- Update your configuration
[gitlab] secret_file
to configure the path to the GitLab secret token. - If you have custom hooks, update your configuration
[hooks] custom_hooks_dir
to configure the path to server-side custom hooks. - Remove the
[gitlab-shell] dir
configuration.
- Update your configuration
16.3.0
-
Update to GitLab 16.3.5 or later. This avoids issue 425971 that causes an excessive use of database disk space for GitLab 16.3.3 and 16.3.4.
-
A unique index was added to ensure that there’s no duplicate NPM packages in the database. If you have duplicate NPM packages, you need to upgrade to 16.1 first, or you are likely to run into the following error:
PG::UniqueViolation: ERROR: could not create unique index "idx_packages_on_project_id_name_version_unique_when_npm"
. -
For Go applications,
crypto/tls
: verifying certificate chains containing large RSA keys is slow (CVE-2023-29409) introduced a hard limit of 8192 bits for RSA keys. In the context of Go applications at GitLab, RSA keys can be configured for:You should check the size of your RSA keys (
openssl rsa -in <your-key-file> -text -noout | grep "Key:"
) for any of the applications above before upgrading. -
A
BackfillCiPipelineVariablesForPipelineIdBigintConversion
background migration is finalized with theEnsureAgainBackfillForCiPipelineVariablesPipelineIdIsFinished
post-deploy migration. GitLab 16.2.0 introduced a batched background migration to backfill bigintpipeline_id
values on theci_pipeline_variables
table. This migration may take a long time to complete on larger GitLab instances (4 hours to process 50 million rows reported in one case). To avoid a prolonged upgrade downtime, make sure the migration has completed successfully before upgrading to 16.3.You can check the size of the
ci_pipeline_variables
table in the database console:select count(*) from ci_pipeline_variables;
Linux package installations
Specific information applies to Linux package installations:
-
In GitLab 16.0, we announced an upgraded base Docker image, which has a new version of OpenSSH Server. An unintended consequence of the new version is that it disables accepting SSH RSA SHA-1 signatures by default. This issue should only impact users using very outdated SSH clients.
To avoid problems with SHA-1 signatures being unavailable, users should update their SSH clients because using SHA-1 signatures is discouraged by the upstream library for security reasons.
To allow for a transition period where users can’t immediately upgrade their SSH clients, GitLab 16.3 and later has support for a
GITLAB_ALLOW_SHA1_RSA
environment variable in theDockerfile
. IfGITLAB_ALLOW_SHA1_RSA
is set totrue
, this deprecated support is reactivated.Because we want to foster security best practices and follow the upstream recommendation, this environment variable will only be available until GitLab 17.0, when we plan to drop support for it.
For more information, see:
- OpenSSH 8.8 release notes.
- An informal explanation.
-
omnibus-gitlab
merge request 7035, which introduces the environment variable.
Geo installations
Specific information applies to installations using Geo:
-
Git pulls against a secondary Geo site are being proxied to the primary Geo site even when that secondary site is up to date. You are impacted if you are using Geo to accelerate remote users who make Git pull requests against a secondary Geo site.
- Impacted versions:
- 16.3.0 to 16.3.3
- Versions containing fix:
- 16.3.4 and later
For more information, see issue 425224.
- Impacted versions:
16.2.0
- Legacy LDAP configuration settings may cause
NoMethodError: undefined method 'devise' for User:Class
errors. This error occurs if you have TLS options (such asca_file
) not specified in thetls_options
hash, or use the legacygitlab_rails['ldap_host']
option. See the configuration workarounds for more details. -
If your GitLab database was created by or upgraded via versions 15.11.0 - 15.11.2 inclusive, upgrading to GitLab 16.2 fails with:
PG::UndefinedColumn: ERROR: column "id_convert_to_bigint" of relation "ci_build_needs" does not exist LINE 1: ...db_config_name:main*/ UPDATE "ci_build_needs" SET "id_conver...
-
You might encounter the following error while upgrading to GitLab 16.2 or later:
main: == 20230620134708 ValidateUserTypeConstraint: migrating ======================= main: -- execute("ALTER TABLE users VALIDATE CONSTRAINT check_0dd5948e38;") rake aborted! StandardError: An error has occurred, all later migrations canceled: PG::CheckViolation: ERROR: check constraint "check_0dd5948e38" of relation "users" is violated by some row
For more information, see issue 421629.
-
You might encounter the following error after upgrading to GitLab 16.2 or later:
PG::NotNullViolation: ERROR: null value in column "source_partition_id" of relation "ci_sources_pipelines" violates not-null constraint
Sidekiq and Puma processes must be restarted to resolve this issue.
Linux package installations
Specific information applies to Linux package installations:
-
As of GitLab 16.2, PostgreSQL 13.11 and 14.8 are both shipped with the Linux package. During a package upgrade, the database isn’t upgraded to PostgreSQL 14. If you want to upgrade to PostgreSQL 14, you must do it manually:
sudo gitlab-ctl pg-upgrade -V 14
PostgreSQL 14 isn’t supported on Geo deployments and is planned for future releases.
-
In 16.2, we are upgrading Redis from 6.2.11 to 7.0.12. This upgrade is expected to be fully backwards compatible.
Redis is not automatically restarted as part of
gitlab-ctl reconfigure
. Hence, users are manually required to runsudo gitlab-ctl restart redis
after the reconfigure run so that the new Redis version gets used. A warning mentioning that the installed Redis version is different than the one running is displayed at the end of reconfigure run until the restart is performed.If your instance has Redis HA with Sentinel, follow the upgrade steps mentioned in Zero Downtime documentation.
Self-compiled installations
- Git 2.41.0 and later is required by Gitaly. You should use the Git version provided by Gitaly.
Geo installations
Specific information applies to installations using Geo:
- New job artifacts are not replicated by Geo if job artifacts are configured to be stored in object storage and
direct_upload
is enabled. This bug is fixed in GitLab versions 16.1.4, 16.2.3, 16.3.0, and later.- Impacted versions: GitLab versions 16.1.0 - 16.1.3 and 16.2.0 - 16.2.2.
- While running an affected version, artifacts which appeared to become synced may actually be missing on the secondary site. Affected artifacts are automatically resynced upon upgrade to 16.1.5, 16.2.5, 16.3.1, 16.4.0, or later. You can manually resync affected job artifacts if needed.
16.1.0
- A
BackfillPreparedAtMergeRequests
background migration is finalized with theFinalizeBackFillPreparedAtMergeRequests
post-deploy migration. GitLab 15.10.0 introduced a batched background migration to backfillprepared_at
values on themerge_requests
table. This migration may take multiple days to complete on larger GitLab instances. Make sure the migration has completed successfully before upgrading to 16.1.0. - GitLab 16.1.0 includes a batched background migration
MarkDuplicateNpmPackagesForDestruction
to mark duplicate NPM packages for destruction. Make sure the migration has completed successfully before upgrading to 16.3.0 or later. -
A
BackfillCiPipelineVariablesForBigintConversion
background migration is finalized with theEnsureBackfillBigintIdIsCompleted
post-deploy migration. GitLab 16.0.0 introduced a batched background migration to backfill bigintid
values on theci_pipeline_variables
table. This migration may take a long time to complete on larger GitLab instances (4 hours to process 50 million rows reported in one case). To avoid a prolonged upgrade downtime, make sure the migration has completed successfully before upgrading to 16.1.You can check the size of the
ci_pipeline_variables
table in the database console:select count(*) from ci_pipeline_variables;
Self-compiled installations
- You must remove any settings related to Puma worker killer from the
puma.rb
configuration file, because those have been removed. For more information, see thepuma.rb.example
file.
Geo installations
Specific information applies to installations using Geo:
- Some project imports do not initialize wiki repositories on project creation. See the details and workaround.
- Because of the migration of project designs to SSF, missing design repositories are being incorrectly flagged as failing verification.
This issue is not a result of an actual replication/verification failure but an invalid internal state for these missing
repositories inside Geo and results in errors in the logs and the verification progress reporting a failed state for
these design repositories. You could be impacted by this issue even if you have not imported projects.
- Impacted versions: GitLab versions 16.1.0 - 16.1.2
- Versions containing fix: GitLab 16.1.3 and later.
- New job artifacts are not replicated by Geo if job artifacts are configured to be stored in object storage and
direct_upload
is enabled. This bug is fixed in GitLab versions 16.1.4, 16.2.3, 16.3.0, and later.- Impacted versions: GitLab versions 16.1.0 - 16.1.3 and 16.2.0 - 16.2.2.
- While running an affected version, artifacts which appeared to become synced may actually be missing on the secondary site. Affected artifacts are automatically resynced upon upgrade to 16.1.5, 16.2.5, 16.3.1, 16.4.0, or later. You can manually resync affected job artifacts if needed.
Wiki repositories not initialized on project creation
Affected minor releases | Affected patch releases | Fixed in |
---|---|---|
15.11 | All | None |
16.0 | All | None |
16.1 | 16.1.0 - 16.1.2 | 16.1.3 and later |
Some project imports do not initialize wiki repositories on project creation. Since the migration of project wikis to SSF, missing wiki repositories are being incorrectly flagged as failing verification. This is not a result of an actual replication/verification failure but an invalid internal state for these missing repositories inside Geo and results in errors in the logs and the verification progress reporting a failed state for these wiki repositories. If you have not imported projects you are not impacted by this issue.
16.0.0
- Sidekiq crashes if there are non-ASCII characters in the
/etc/gitlab/gitlab.rb
file. You can fix this by following the workaround in issue 412767. - Sidekiq jobs are only routed to
default
andmailers
queues by default, and as a result, every Sidekiq process also listens to those queues to ensure all jobs are processed across all queues. This behavior does not apply if you have configured the routing rules. - Docker 20.10.10 or later is required to run the GitLab Docker image. Older versions throw errors on startup.
- Starting with 16.0, GitLab self-managed installations now have two database connections by default, instead of one. This change doubles the number of PostgreSQL connections. It makes self-managed versions of GitLab behave similarly to GitLab.com, and is a step toward enabling a separate database for CI features for self-managed versions of GitLab. Before upgrading to 16.0, determine if you need to increase max connections for PostgreSQL.
- This change applies to installation methods with Linux packages (Omnibus), GitLab Helm chart, GitLab Operator, GitLab Docker images, and self-compiled installations.
Linux package installations
Specific information applies to Linux package installations:
-
The binaries for PostgreSQL 12 have been removed.
Prior to upgrading, administrators of Linux package installations must ensure the installation is using PostgreSQL 13.
- Grafana that was bundled with GitLab is deprecated and is no longer supported. It is removed in GitLab 16.3.
-
This upgrades
openssh-server
to1:8.9p1-3
.Using
ssh-keyscan -t rsa
with older OpenSSH clients to obtain public key information is no longer viable because of the deprecations listed in OpenSSH 8.7 Release Notes.Workaround is to make use of a different key type, or upgrade the client OpenSSH to a version >= 8.7.
-
Migrate your Praefect configuration to the new structure to ensure all your
praefect['..']
settings continue to work in GitLab 16.0 and later. -
Migrate your Gitaly configuration to the new structure
to ensure all your
gitaly['..']
settings continue to work in GitLab 16.0 and later.
Geo installations
Specific information applies to installations using Geo:
- Some project imports do not initialize wiki repositories on project creation. See the details and workaround.
Gitaly configuration structure change
The Gitaly configuration structure in the Linux package changes in GitLab 16.0 to be consistent with the Gitaly configuration structure used in self-compiled installations.
As a result of this change, a single hash under gitaly['configuration']
holds most Gitaly
configuration. Some gitaly['..']
configuration options continue to be used by GitLab 16.0 and later:
enable
dir
bin_path
env_directory
env
open_files_ulimit
consul_service_name
consul_service_meta
Migrate by moving your existing configuration under the new structure. git_data_dirs
is supported until GitLab 17.0. The new structure is supported from GitLab 15.10.
Migrate to the new structure
- If you’re running Gitaly Cluster, ensure repository verification is disabled on all Praefect nodes.
Configure
verification_interval: 0
, and apply withgitlab-ctl reconfigure
. - When applying the new structure to your configuration
- Replace the
...
with the value from the old key. - When configuring
storage
to replacegit_data_dirs
, you must appendrepositories
to the path as documented below. If you miss this out your Git repositories are inaccessible until the configuration is fixed. This misconfiguration can cause metadata deletion, and is the reason for disabling repository verification. - Skip any keys you haven’t configured a value for previously.
- Recommended. Include a trailing comma for all hash keys so the hash remains valid when keys are re-ordered or additional keys are added.
- Replace the
- Apply the change with
gitlab-ctl reconfigure
. - Test Git repository functionality in GitLab.
- Remove the old keys from the configuration once migrated, and then re-run
gitlab-ctl reconfigure
. - Recommended, if you’re running Gitaly Cluster. Reinstate Praefect repository verification
by removing
verification_interval: 0
.
The new structure is documented below with the old keys described in a comment above the new keys.
gitaly['configuration'] = {
# gitaly['socket_path']
socket_path: ...,
# gitaly['runtime_dir']
runtime_dir: ...,
# gitaly['listen_addr']
listen_addr: ...,
# gitaly['prometheus_listen_addr']
prometheus_listen_addr: ...,
# gitaly['tls_listen_addr']
tls_listen_addr: ...,
tls: {
# gitaly['certificate_path']
certificate_path: ...,
# gitaly['key_path']
key_path: ...,
},
# gitaly['graceful_restart_timeout']
graceful_restart_timeout: ...,
logging: {
# gitaly['logging_level']
level: ...,
# gitaly['logging_format']
format: ...,
# gitaly['logging_sentry_dsn']
sentry_dsn: ...,
# gitaly['logging_ruby_sentry_dsn']
ruby_sentry_dsn: ...,
# gitaly['logging_sentry_environment']
sentry_environment: ...,
# gitaly['log_directory']
dir: ...,
},
prometheus: {
# gitaly['prometheus_grpc_latency_buckets']. The old value was configured as a string
# such as '[0, 1, 2]'. The new value must be an array like [0, 1, 2].
grpc_latency_buckets: ...,
},
auth: {
# gitaly['auth_token']
token: ...,
# gitaly['auth_transitioning']
transitioning: ...,
},
git: {
# gitaly['git_catfile_cache_size']
catfile_cache_size: ...,
# gitaly['git_bin_path']
bin_path: ...,
# gitaly['use_bundled_git']
use_bundled_binaries: ...,
# gitaly['gpg_signing_key_path']
signing_key: ...,
# gitaly['gitconfig']. This is still an array but the type of the elements have changed.
config: [
{
# Previously the elements contained 'section', and 'subsection' in addition to 'key'. Now
# these all should be concatenated into just 'key', separated by dots. For example,
# {section: 'first', subsection: 'middle', key: 'last', value: 'value'}, should become
# {key: 'first.middle.last', value: 'value'}.
key: ...,
value: ...,
},
],
},
# Storage could previously be configured through either gitaly['storage'] or 'git_data_dirs'. Migrate
# the relevant configuration according to the instructions below.
# For 'git_data_dirs', migrate only the 'path' to the gitaly['configuration'] and leave the rest of it untouched.
storage: [
{
# gitaly['storage'][<index>]['name']
#
# git_data_dirs[<name>]. The storage name was configured as a key in the map.
name: ...,
# gitaly['storage'][<index>]['path']
#
# git_data_dirs[<name>]['path']. Use the value from git_data_dirs[<name>]['path'] and append '/repositories' to it.
#
# For example, if the path in 'git_data_dirs' was '/var/opt/gitlab/git-data', use
# '/var/opt/gitlab/git-data/repositories'. The '/repositories' extension was automatically
# appended to the path configured in `git_data_dirs`.
path: ...,
},
],
hooks: {
# gitaly['custom_hooks_dir']
custom_hooks_dir: ...,
},
daily_maintenance: {
# gitaly['daily_maintenance_disabled']
disabled: ...,
# gitaly['daily_maintenance_start_hour']
start_hour: ...,
# gitaly['daily_maintenance_start_minute']
start_minute: ...,
# gitaly['daily_maintenance_duration']
duration: ...,
# gitaly['daily_maintenance_storages']
storages: ...,
},
cgroups: {
# gitaly['cgroups_mountpoint']
mountpoint: ...,
# gitaly['cgroups_hierarchy_root']
hierarchy_root: ...,
# gitaly['cgroups_memory_bytes']
memory_bytes: ...,
# gitaly['cgroups_cpu_shares']
cpu_shares: ...,
repositories: {
# gitaly['cgroups_repositories_count']
count: ...,
# gitaly['cgroups_repositories_memory_bytes']
memory_bytes: ...,
# gitaly['cgroups_repositories_cpu_shares']
cpu_shares: ...,
}
},
# gitaly['concurrency']. While the structure is the same, the string keys in the array elements
# should be replaced by symbols as elsewhere. {'key' => 'value'}, should become {key: 'value'}.
concurrency: ...,
# gitaly['rate_limiting']. While the structure is the same, the string keys in the array elements
# should be replaced by symbols as elsewhere. {'key' => 'value'}, should become {key: 'value'}.
rate_limiting: ...,
pack_objects_cache: {
# gitaly['pack_objects_cache_enabled']
enabled: ...,
# gitaly['pack_objects_cache_dir']
dir: ...,
# gitaly['pack_objects_cache_max_age']
max_age: ...,
}
}
Praefect configuration structure change
The Praefect configuration structure in the Linux package changes in GitLab 16.0 to be consistent with the Praefect configuration structure used in self-compiled installations.
As a result of this change, a single hash under praefect['configuration']
holds most Praefect
configuration. Some praefect['..']
configuration options continue to be used by GitLab 16.0 and later:
enable
dir
log_directory
env_directory
env
wrapper_path
auto_migrate
consul_service_name
Migrate by moving your existing configuration under the new structure. The new structure is supported from GitLab 15.9.
Migrate to the new structure
- When applying the new structure to your configuration:
- Replace the
...
with the value from the old key. - Disable repository verification using
verification_interval: 0
, as shown below. - Skip any keys you haven’t configured a value for previously.
- Recommended. Include a trailing comma for all hash keys so the hash remains valid when keys are re-ordered or additional keys are added.
- Replace the
- Apply the change with
gitlab-ctl reconfigure
. - Test Git repository functionality in GitLab.
- Remove the old keys from the configuration once migrated, and then re-run
gitlab-ctl reconfigure
.
The new structure is documented below with the old keys described in a comment above the new keys.
praefect['configuration'] = {
# praefect['listen_addr']
listen_addr: ...,
# praefect['socket_path']
socket_path: ...,
# praefect['prometheus_listen_addr']
prometheus_listen_addr: ...,
# praefect['tls_listen_addr']
tls_listen_addr: ...,
# praefect['separate_database_metrics']
prometheus_exclude_database_from_default_metrics: ...,
auth: {
# praefect['auth_token']
token: ...,
# praefect['auth_transitioning']
transitioning: ...,
},
logging: {
# praefect['logging_format']
format: ...,
# praefect['logging_level']
level: ...,
},
failover: {
# praefect['failover_enabled']
enabled: ...,
},
background_verification: {
# praefect['background_verification_delete_invalid_records']
delete_invalid_records: ...,
# praefect['background_verification_verification_interval']
#
# IMPORTANT:
# As part of reconfiguring Praefect, disable this feature.
# Read about this above.
#
verification_interval: 0,
},
reconciliation: {
# praefect['reconciliation_scheduling_interval']
scheduling_interval: ...,
# praefect['reconciliation_histogram_buckets']. The old value was configured as a string
# such as '[0, 1, 2]'. The new value must be an array like [0, 1, 2].
histogram_buckets: ...,
},
tls: {
# praefect['certificate_path']
certificate_path: ...,
# praefect['key_path']
key_path: ...,
},
database: {
# praefect['database_host']
host: ...,
# praefect['database_port']
port: ...,
# praefect['database_user']
user: ...,
# praefect['database_password']
password: ...,
# praefect['database_dbname']
dbname: ...,
# praefect['database_sslmode']
sslmode: ...,
# praefect['database_sslcert']
sslcert: ...,
# praefect['database_sslkey']
sslkey: ...,
# praefect['database_sslrootcert']
sslrootcert: ...,
session_pooled: {
# praefect['database_direct_host']
host: ...,
# praefect['database_direct_port']
port: ...,
# praefect['database_direct_user']
user: ...,
# praefect['database_direct_password']
password: ...,
# praefect['database_direct_dbname']
dbname: ...,
# praefect['database_direct_sslmode']
sslmode: ...,
# praefect['database_direct_sslcert']
sslcert: ...,
# praefect['database_direct_sslkey']
sslkey: ...,
# praefect['database_direct_sslrootcert']
sslrootcert: ...,
}
},
sentry: {
# praefect['sentry_dsn']
sentry_dsn: ...,
# praefect['sentry_environment']
sentry_environment: ...,
},
prometheus: {
# praefect['prometheus_grpc_latency_buckets']. The old value was configured as a string
# such as '[0, 1, 2]'. The new value must be an array like [0, 1, 2].
grpc_latency_buckets: ...,
},
# praefect['graceful_stop_timeout']
graceful_stop_timeout: ...,
# praefect['virtual_storages']. The old value was a hash map but the new value is an array.
virtual_storage: [
{
# praefect['virtual_storages'][VIRTUAL_STORAGE_NAME]. The name was previously the key in
# the 'virtual_storages' hash.
name: ...,
# praefect['virtual_storages'][VIRTUAL_STORAGE_NAME]['nodes'][NODE_NAME]. The old value was a hash map
# but the new value is an array.
node: [
{
# praefect['virtual_storages'][VIRTUAL_STORAGE_NAME]['nodes'][NODE_NAME]. Use NODE_NAME key as the
# storage.
storage: ...,
# praefect['virtual_storages'][VIRTUAL_STORAGE_NAME]['nodes'][NODE_NAME]['address'].
address: ...,
# praefect['virtual_storages'][VIRTUAL_STORAGE_NAME]['nodes'][NODE_NAME]['token'].
token: ...,
},
],
}
]
}
Long-running user type data change
GitLab 16.0 is a required stop for large GitLab instances with a lot of records in the users
table.
The threshold is 30,000 users, which includes:
- Developers and other users in any state, including active, blocked, and pending approval.
- Bot accounts for project and group access tokens.
GitLab 16.0 introduced a batched background migration to
migrate user_type
values from NULL
to 0
. This
migration might take multiple days to complete on larger GitLab instances. Make sure the migration
has completed successfully before upgrading to 16.1.0 or later.
GitLab 16.1 introduces the FinalizeUserTypeMigration
migration which ensures the
16.0 MigrateHumanUserType
background migration is completed, executing the 16.0 change synchronously
during the upgrade if it’s not completed.
GitLab 16.2 implements a NOT NULL
database constraint
which fails if the 16.0 migration is not complete.
If 16.0 has been skipped (or the 16.0 migration is not complete) subsequent Linux package (Omnibus) and Docker upgrades might fail after an hour:
FATAL: Mixlib::ShellOut::CommandTimeout: rails_migration[gitlab-rails]
[..]
Mixlib::ShellOut::CommandTimeout: Command timed out after 3600s:
There is a fix-forward workaround for this issue.
While the workaround is completing the database changes, GitLab is likely to be in
an unusable state, generating 500
errors. The errors are caused by Sidekiq and Puma running
application code that is incompatible with the database schema.
At the end of the workaround process, Sidekiq and Puma are restarted to resolve that issue.
Undefined column error upgrading to 16.2 or later
A bug in GitLab 15.11 incorrectly disabled a database change on self-managed instances. For more information, see issue 408835.
If your GitLab instance upgraded first to 15.11.0, 15.11.1, or 15.11.2 the database schema is incorrect and upgrading to GitLab 16.2 or later fails with an error. A database change requires the earlier modification to be in place:
PG::UndefinedColumn: ERROR: column "id_convert_to_bigint" of relation "ci_build_needs" does not exist
LINE 1: ...db_config_name:main*/ UPDATE "ci_build_needs" SET "id_conver...
GitLab 15.11.3 shipped a fix for this bug, but it doesn’t correct the problem on instances already running the earlier 15.11 releases.
If you’re not sure if an instance is affected, check for the column on the database console:
select pg_typeof (id_convert_to_bigint) from public.ci_build_needs limit 1;
If you need the workaround, this query fails:
ERROR: column "id_convert_to_bigintd" does not exist
LINE 1: select pg_typeof (id_convert_to_bigintd) from public.ci_buil...
Unaffected instances return:
pg_typeof
-----------
bigint
The workaround for this issue differs if your GitLab instance’s database schema was recently created:
Installation version | Workaround |
---|---|
15.9 or earlier | 15.9 |
15.10 | 15.10 |
15.11 | 15.11 |
Most instances should use the 15.9 procedure. Only very new instances require the the 15.10 or 15.11 procedures. If you’ve migrated GitLab using backup and restore, the database schema comes from the original instance. Select the workaround based on the source instance.
The commands in the following sections are for Linux package installations, and differ for other installation types:
- Omit
sudo
-
Shell into the GitLab container and run the same commands:
docker exec -it <container-id> bash
- Use
sudo -u git -H bundle exec rake RAILS_ENV=production
instead ofsudo gitlab-rake
- Run the SQL on your PostgreSQL database console
- Omit
sudo
. - Shell into the
toolbox
pod to run the Rake commands:gitlab-rake
is in/usr/local/bin
if not in thePATH
.- Refer to our Kubernetes cheat sheet for details.
- Run the SQL on your PostgreSQL database console
Workaround: instance created with 15.9 or earlier
# Restore schema
sudo gitlab-psql -c "DELETE FROM schema_migrations WHERE version IN ('20230130175512', '20230130104819');"
sudo gitlab-rake db:migrate:up VERSION=20230130175512
sudo gitlab-rake db:migrate:up VERSION=20230130104819
# Re-schedule background migrations
sudo gitlab-rake db:migrate:down VERSION=20230130202201
sudo gitlab-rake db:migrate:down VERSION=20230130110855
sudo gitlab-rake db:migrate:up VERSION=20230130202201
sudo gitlab-rake db:migrate:up VERSION=20230130110855
Workaround: instance created with 15.10
# Restore schema for sent_notifications
sudo gitlab-psql -c "DELETE FROM schema_migrations WHERE version = '20230130175512';"
sudo gitlab-rake db:migrate:up VERSION=20230130175512
# Re-schedule background migration for sent_notifications
sudo gitlab-rake db:migrate:down VERSION=20230130202201
sudo gitlab-rake db:migrate:up VERSION=20230130202201
# Restore schema for ci_build_needs
sudo gitlab-rake db:migrate:down VERSION=20230321163547
sudo gitlab-psql -c "INSERT INTO schema_migrations (version) VALUES ('20230321163547');"
Workaround: instance created with 15.11
# Restore schema for sent_notifications
sudo gitlab-rake db:migrate:down VERSION=20230411153310
sudo gitlab-psql -c "INSERT INTO schema_migrations (version) VALUES ('20230411153310');"
# Restore schema for ci_build_needs
sudo gitlab-rake db:migrate:down VERSION=20230321163547
sudo gitlab-psql -c "INSERT INTO schema_migrations (version) VALUES ('20230321163547');"