Cases6 cases

3 to 6 of 6

1. IL-0216 Wikimedia on-wiki search was slow or rejected after a datacenter switchover overloaded one Elasticsearch node in codfw

   Wikimedia Foundation2023

   Symptoms

   Right after the primary datacenter was switched from EQIAD to CODFW at 14:00 UTC on 20 September 2023, on-wiki search on all Wikimedia wikis became delayed or unavailable for a subset of users. At 14:08 the alert "CirrusSearch codfw 95th percentile latency - more_like" fired and reached IRC and the Search Platform email list as "PROBLEM alert - graphite1005/CirrusSearch codfw 95th percentile latency - more_like is CRITICAL".

   High latencyHigh error rateLoad imbalanceMisrouted requests

   Trigger

   At 14:00 UTC the primary datacenter was switched from EQIAD to CODFW. more_like queries that were expected to go to eqiad, the non-primary datacenter, went to codfw instead, apparently because RESTBase sent them as POST rather than GET.

   Root cause

   Software defect

   Component

   Elasticsearch or OpenSearchIn-memory cache

   Contributing factor

   Unverified assumptionDependence on a warm cacheNo runbook

   Detection

   Automated monitoring caught the problem, but not quickly enough to prevent user-facing impact: the Grafana alert "CirrusSearch codfw 95th percentile latency - more_like" went to IRC and the Search Platform email list at 14:08. No one was paged; SREs were already watching closely because of the datacenter switchover. At 14:14 Hashar pinged #wikimedia-search and engineers began troubleshooting.

   Remediation

   • Banned the overloaded node elastic2044 from the Elasticsearch cluster

   Action items

   • Work out which Elasticsearch steps are needed before a datacenter switchover
   • Update the Elasticsearch section of the datacenter switchover documentation
   • Review the suggested switchover improvements in T285347 and decide whether to implement them
   • Make the RESTBase /related/{article} endpoint send GET instead of POST so the CirrusSearch query cache stays warm in both datacenters
   • Review the shard balance of busy wikis, especially enwiki_content

   Lessons

   Monitoring caught the problem, but not quickly enough to prevent user-facing impact. The heightened attention during the switchover meant the impact was recognised immediately, and banning a single node was quick and easy. The mitigation was not guided by an existing runbook, and the shards on the overloaded host were not checked before it was banned.

   View case
2. IL-0217 Honeycomb's main MySQL database locked up under cache-miss reads and ran out of connections, a 68-minute total outage

   Honeycomb2023

   Symptoms

   Hours after a routine switch between two clusters of Retriever, Honeycomb's storage and query engine, late on 24 July 2023, the internal SLO for Shepherd, the ingest service, began to burn slowly. The performance was only marginally slow and inconsistent, with no good explanation, so engineers treated it as minor and left it for the morning.

   High latencyDatabase overloadLock contention or deadlockService outage

   Trigger

   Late on 24 July Honeycomb switched between two Retriever clusters to avoid a subtle bug, and the new cluster stopped writing the last-written timestamps the schema cache depends on. On the morning of 25 July a feature flag flip meant to move those writes back stopped the remaining writes entirely.

   Root cause

   Software defect

   Component

   DatabaseIn-memory cache

   Contributing factor

   Dependence on a warm cacheIgnored early signalHarmful manual actionUnverified assumption

   Detection

   Hours after the cluster switch, the internal SLO for the ingest service Shepherd started burning slowly. Engineers left it for the morning, when a chart showed that Retriever's database calls updating the last-written records had stopped.

   Remediation

   • Flipped only the writes back to the other Retriever cluster with a feature flag, which did not fix the problem
   • Set up circuit breakers that denied all Shepherd traffic with a 5xx error to protect the database
   • Failed the hard-locked database over to a replica
   • Restarted all Retriever hosts so they could provide data for the cache
   • Set the last-written timestamp of every schema with traffic in the last day to now, so the schema store reloaded the cache
   • Removed the circuit breaker and restarted the query hosts that failed internal checks

   Action items

   • Finish the Retriever cluster migration and remove all code that could disable writes behind the cache
   • Explore ways to strengthen the cache further
   • Lower the contention put on the database during update storms from new schemas
   • Stabilise the performance costs of some operations in the system
   • Study instrumentation and experimentation that detect and handle such edge cases better

   Lessons

   Locally good decisions combined into the outage: the query engine wrote the timestamps because it already saw all the data, the shared cache helped scale ingest and untangle the main database, the cluster switch was done to avoid a bug, frequent deploys hid the feature flag bug, pausing deploys during the investigation kept hosts from restarting, and past near-misses made the team expect a slow failover. Feature flags, frequent deploys, pausing deploys during incidents and learning from near-misses all contributed, although they are among the practices Honeycomb trusts most. For a rare lockup of this kind a database failover is the fastest recovery. Several recent Honeycomb incidents started from trying to avoid a subtle bug.

   View case
3. IL-0212 Many Slack users could not connect because a Vitess keyspace for channel membership was overloaded by cache-miss queries

   Slack2022

   Symptoms

   Just after 6 a.m. Pacific Time on 22 February 2022, user tickets about problems connecting to Slack, problems reported by internal users, and pages and alerts to several engineering teams all arrived at almost the same moment. Many users could not connect: the Slack client could not finish booting, the step that fetches channel lists, preferences and recent conversations, and showed an error page.

   Database overloadHigh error rateHigh latency

   Trigger

   A percentage-based rollout step upgraded the Consul agent on 25% of the fleet, followed by a slow sequential restart of the agents. Each restart on a Memcached node removed it from the Consul service catalog, so Mcrib, the cache-tier control plane, replaced it with an empty spare or flushed it when it rejoined, and the cache hit rate dropped as Slack reached its daily peak traffic.

   Root cause

   Capacity exhaustion

   Component

   DatabaseIn-memory cacheService discovery

   Contributing factor

   Dependence on a warm cacheInefficient queryAutomation amplificationRetry stormChange during peak hours

   Detection

   User tickets about connection problems, reports from internal users and pages and alerts to several engineering teams arrived almost at the same time, just after 6 a.m. Pacific Time.

   Remediation

   • Paused the Consul agent restart operation
   • Throttled client boot requests to reduce load on the database tier
   • Cut the client boot rate limit again after a too-large increase and then raised it in smaller steps
   • Changed the GDM membership query to read from Vitess only the data missing from Memcached
   • Let the query read from replicas as well as Vitess primaries
   • Raised the client boot rate limit slowly back to normal traffic levels as the caches refilled

   Action items

   • Change the process for Consul rollouts so upgrades do not trigger this failure mode again
   • Change the Mcrib control loop so a brief network outage or partition of cache nodes does not set off the same cascading failure
   • Make the GDM membership query read from a table sharded by channel
   • Check the other cache-fronted database queries for high-volume scatter queries
   • Explore longer-term ways to make the caching tier more resilient

   Lessons

   Slack describes performance bottlenecks, a cascading failure, dependence on a warm cache and a change as the trigger. Once the system had passed the tipping point, lower cache hit rates caused high database load that in turn kept the cache from filling, and a system in that state stays there until an outside intervention, a large cut in load or an increase in capacity, changes it. Mcrib contributed not by being wrong but by being faster than the old Consul lock scheme at replacing Memcached nodes, which increased churn in the cache tier. Client retries still added load despite backoff and jitter. A brief network partition of the cache nodes could set off the same failure, and the interaction between Consul and the caching architecture is now more widely understood.

   View case
4. IL-0297 Mezmo users could not sign in and log indexing lagged after the auth cache's Redis servers were OOM killed

   Mezmo2019

   Symptoms

   On 15 May 2019, alerts and customer reports showed that the Mezmo web application did not let users sign in, and the status page reported an indexing delay: incoming logs reached live tail and search late.

   Service outageOOMKilledDatabase overloadThroughput degradation

   Trigger

   A traffic spike that the tool tracking batches of logs was not sized to handle.

   Root cause

   Capacity exhaustion

   Component

   In-memory cacheDatabase

   Contributing factor

   Dependence on a warm cache

   Detection

   Alerts and customer reports in the early morning hours triggered the escalation procedure and woke the on-call engineers.

   Remediation

   • Resized specific parts of the cluster to handle a much higher load

   Action items

   • Retool the batch tracking system to reduce its direct impact on the Redis clusters
   • Add observability instrumentation around the batch tracking feature set
   • Improve monitoring for Redis
   • Remove the bottlenecks in the pipeline that were involved in this incident

   Lessons

   Mezmo took from the incident that deep instrumentation, tracing and observability across the platform matter more as features scale up.

   View case

By root cause4