Monitoring Gitaly and Gitaly Cluster

You can use the available logs and Prometheus metrics to monitor Gitaly and Gitaly Cluster (Praefect).

Metric definitions are available:

  • Directly from Prometheus /metrics endpoint configured for Gitaly.
  • Using Grafana Explore on a Grafana instance configured against Prometheus.

Monitor Gitaly rate limiting

Gitaly can be configured to limit requests based on:

  • Concurrency of requests.
  • A rate limit.

Monitor Gitaly request limiting with the gitaly_requests_dropped_total Prometheus metric. This metric provides a total count of requests dropped due to request limiting. The reason label indicates why a request was dropped:

  • rate, due to rate limiting.
  • max_size, because the concurrency queue size was reached.
  • max_time, because the request exceeded the maximum queue wait time as configured in Gitaly.

Monitor Gitaly concurrency limiting

You can observe specific behavior of concurrency-queued requests using Gitaly logs and Prometheus.

In the Gitaly logs, you can identify logs related to the pack-objects concurrency limiting with entries such as:

Log FieldDescription
limit.concurrency_queue_lengthIndicates the current length of the queue specific to the RPC type of the ongoing call. It provides insight into the number of requests waiting to be processed due to concurrency limits.
limit.concurrency_queue_msRepresents the duration, in milliseconds, that a request has spent waiting in the queue due to the limit on concurrent RPCs. This field helps understand the impact of concurrency limits on request processing times.
limit.concurrency_droppedIf the request is dropped due to limits being reached, this field specifies the reason: either max_time (request waited in the queue longer than the maximum allowed time) or max_size (the queue reached its maximum size).
limit.limiting_keyIdentifies the key used for limiting.
limit.limiting_typeSpecifies the type of process being limited. In this context, it’s per-rpc, indicating that the concurrency limiting is applied on a per-RPC basis.

For example:

{
  "limit .concurrency_queue_length": 1,
  "limit .concurrency_queue_ms": 0,
  "limit.limiting_key": "@hashed/79/02/7902699be42c8a8e46fbbb450172651786b22c56a189f7625a6da49081b2451.git",
  "limit.limiting_type": "per-rpc"
}

In Prometheus, look for the following metrics:

  • gitaly_concurrency_limiting_in_progress indicates how many concurrent requests are being processed.
  • gitaly_concurrency_limiting_queued indicates how many requests for an RPC for a given repository are waiting due to the concurrency limit being reached.
  • gitaly_concurrency_limiting_acquiring_seconds indicates how long a request has to wait due to concurrency limits before being processed.

Monitor Gitaly pack-objects concurrency limiting

You can observe specific behavior of pack-objects limiting using Gitaly logs and Prometheus.

In the Gitaly logs, you can identify logs related to the pack-objects concurrency limiting with entries such as:

Log FieldDescription
limit.concurrency_queue_lengthCurrent length of the queue for the pack-objects processes. Indicates the number of requests that are waiting to be processed because the limit on concurrent processes has been reached.
limit.concurrency_queue_msTime a request has spent waiting in the queue, in milliseconds. Indicates how long a request has had to wait because of the limits on concurrency.
limit.limiting_keyRemote IP of the sender.
limit.limiting_typeType of process being limited. In this case, pack-objects.

Example configuration:

{
  "limit .concurrency_queue_length": 1,
  "limit .concurrency_queue_ms": 0,
  "limit.limiting_key": "1.2.3.4",
  "limit.limiting_type": "pack-objects"
}

In Prometheus, look for the following metrics:

  • gitaly_pack_objects_in_progress indicates how many pack-objects processes are being processed concurrently.
  • gitaly_pack_objects_queued indicates how many requests for pack-objects processes are waiting due to the concurrency limit being reached.
  • gitaly_pack_objects_acquiring_seconds indicates how long a request for a pack-object process has to wait due to concurrency limits before being processed.

Monitor Gitaly cgroups

You can observe the status of control groups (cgroups) using Prometheus:

  • gitaly_cgroups_reclaim_attempts_total, a gauge for the total number of times there has been a memory reclaim attempt. This number resets each time a server is restarted.
  • gitaly_cgroups_cpu_usage, a gauge that measures CPU usage per cgroup.
  • gitaly_cgroup_procs_total, a gauge that measures the total number of processes Gitaly has spawned under the control of cgroups.
  • gitaly_cgroup_cpu_cfs_periods_total, a counter that for the value of nr_periods.
  • gitaly_cgroup_cpu_cfs_throttled_periods_total, a counter for the value of nr_throttled.
  • gitaly_cgroup_cpu_cfs_throttled_seconds_total, a counter for the value of throttled_time in seconds.

pack-objects cache

The following pack-objects cache metrics are available:

  • gitaly_pack_objects_cache_enabled, a gauge set to 1 when the cache is enabled. Available labels: dir and max_age.
  • gitaly_pack_objects_cache_lookups_total, a counter for cache lookups. Available label: result.
  • gitaly_pack_objects_generated_bytes_total, a counter for the number of bytes written into the cache.
  • gitaly_pack_objects_served_bytes_total, a counter for the number of bytes read from the cache.
  • gitaly_streamcache_filestore_disk_usage_bytes, a gauge for the total size of cache files. Available label: dir.
  • gitaly_streamcache_index_entries, a gauge for the number of entries in the cache. Available label: dir.

Some of these metrics start with gitaly_streamcache because they are generated by the streamcache internal library package in Gitaly.

Example:

gitaly_pack_objects_cache_enabled{dir="/var/opt/gitlab/git-data/repositories/+gitaly/PackObjectsCache",max_age="300"} 1
gitaly_pack_objects_cache_lookups_total{result="hit"} 2
gitaly_pack_objects_cache_lookups_total{result="miss"} 1
gitaly_pack_objects_generated_bytes_total 2.618649e+07
gitaly_pack_objects_served_bytes_total 7.855947e+07
gitaly_streamcache_filestore_disk_usage_bytes{dir="/var/opt/gitlab/git-data/repositories/+gitaly/PackObjectsCache"} 2.6200152e+07
gitaly_streamcache_filestore_removed_total{dir="/var/opt/gitlab/git-data/repositories/+gitaly/PackObjectsCache"} 1
gitaly_streamcache_index_entries{dir="/var/opt/gitlab/git-data/repositories/+gitaly/PackObjectsCache"} 1

Queries

The following are some queries for monitoring Gitaly:

  • Use the following Prometheus query to observe the type of connections Gitaly is serving a production environment:

    sum(rate(gitaly_connections_total[5m])) by (type)
    
  • Use the following Prometheus query to monitor the authentication behavior of your GitLab installation:

    sum(rate(gitaly_authentications_total[5m])) by (enforced, status)
    

    In a system where authentication is configured correctly and where you have live traffic, you see something like this:

    {enforced="true",status="ok"}  4424.985419441742
    

    There may also be other numbers with rate 0, but you only have to take note of the non-zero numbers.

    The only non-zero number should have enforced="true",status="ok". If you have other non-zero numbers, something is wrong in your configuration.

    The status="ok" number reflects your current request rate. In the example above, Gitaly is handling about 4000 requests per second.

  • Use the following Prometheus query to observe the Git protocol versions being used in a production environment:

    sum(rate(gitaly_git_protocol_requests_total[1m])) by (grpc_method,git_protocol,grpc_service)
    

Monitor Gitaly Cluster

To monitor Gitaly Cluster (Praefect), you can use these Prometheus metrics. Two separate metrics endpoints are available from which metrics can be scraped:

  • The default /metrics endpoint.
  • /db_metrics, which contains metrics that require database queries.

Default Prometheus /metrics endpoint

The following metrics are available from the /metrics endpoint:

  • gitaly_praefect_read_distribution, a counter to track distribution of reads. It has two labels:

    • virtual_storage.
    • storage.

    They reflect configuration defined for this instance of Praefect.

  • gitaly_praefect_replication_latency_bucket, a histogram measuring the amount of time it takes for replication to complete after the replication job starts. Available in GitLab 12.10 and later.
  • gitaly_praefect_replication_delay_bucket, a histogram measuring how much time passes between when the replication job is created and when it starts. Available in GitLab 12.10 and later.
  • gitaly_praefect_connections_total, the total number of connections to Praefect. Introduced in GitLab 14.7.
  • gitaly_praefect_method_types, a count of accessor and mutator RPCs per node.

To monitor strong consistency, you can use the following Prometheus metrics:

  • gitaly_praefect_transactions_total, the number of transactions created and voted on.
  • gitaly_praefect_subtransactions_per_transaction_total, the number of times nodes cast a vote for a single transaction. This can happen multiple times if multiple references are getting updated in a single transaction.
  • gitaly_praefect_voters_per_transaction_total: the number of Gitaly nodes taking part in a transaction.
  • gitaly_praefect_transactions_delay_seconds, the server-side delay introduced by waiting for the transaction to be committed.
  • gitaly_hook_transaction_voting_delay_seconds, the client-side delay introduced by waiting for the transaction to be committed.

To monitor repository verification, use the following Prometheus metrics:

  • gitaly_praefect_verification_jobs_dequeued_total, the number of verification jobs picked up by the worker.
  • gitaly_praefect_verification_jobs_completed_total, the number of verification jobs completed by the worker. The result label indicates the end result of the jobs:
    • valid indicates the expected replica existed on the storage.
    • invalid indicates the replica expected to exist did not exist on the storage.
    • error indicates the job failed and has to be retried.
  • gitaly_praefect_stale_verification_leases_released_total, the number of stale verification leases released.

You can also monitor the Praefect logs.

Database metrics /db_metrics endpoint

Introduced in GitLab 14.5.

The following metrics are available from the /db_metrics endpoint:

  • gitaly_praefect_unavailable_repositories, the number of repositories that have no healthy, up to date replicas.
  • gitaly_praefect_replication_queue_depth, the number of jobs in the replication queue.
  • gitaly_praefect_verification_queue_depth, the total number of replicas pending verification.
  • gitaly_praefect_read_only_repositories, the number of repositories in read-only mode in a virtual storage.
    • This metric was removed in GitLab 15.4.