Create your own GitHub profile
Sign up for your own profile on GitHub, the best place to host code, manage projects, and build software alongside 50 million developers.
Sign up
Pinned
1,586 contributions in the last year
Contribution activity
October 2020
Created a pull request in vespa-engine/vespa that received 2 comments
Rebalance nodes to exclusive switches
SwitchRebalancer drives the balancing using the same mechanism as the existing
Rebalancer.
Node movers may second-guess each other, but that appear…
- Reduce locking in LoadBalancerExpirer
- Take correct lock for each node when retiring
- NodeList cleanup
- Hide shared endpoints when feature flag is set
- Hold lock while committing transaction
- Reduce number of mock provisioners
- Use lock default timeout in NodeFailer
- Acquire provision lock for transactional writes
- Reduce LoadBalancerExpirer lock timeout
- Remove config lock from node-repository
- Remove unused weighted-dns-per-region flag
- Stop taking config lock exclusively in node-repository
- Revert "Take both application locks during transition"
- Remove node-object-cache-size flag
- Remove unused Utf8 methods
- Take both application locks during transition
- Remove custom Utf8.toBytes implementation
- Remove custom Utf8.toString implementation
- Store correct agent when adding nodes
- Use deploy timeout in config server to set reservation expiry
- Deadlock detection
- Hide global non-direct endpoints if configured
- Record quota usage
- Do not make assumptions on presence of provisioned
- Avoid metrics on reentry of lock
- Remove enable-crowdstrike, -nessus, and -fleet-sshd-config flags
- Avoid even small double-counting of locked time
- Set stack trace on correct JSON object
- Revert "Remove unused Utf8 methods"
- Move lock metrics to MetricsReporter
- Remove unused throttle policy override
- Don't reactivate on retired hosts
- Quota as BigDecimals in the controller
- Add metrics to lock attempts
- Record locks taken for external deploys