[tor-project] minutes from the sysadmins

Roll call: who’s there and emergencies

anarcat, groente, lavamind, lelutin and zen

Dashboard review

Normal per-user check-in:

General dashboards:

FYI: tpo/tpa/tails/sysadmin moved to tpo/tpa/tails-sysadmin

Just that.

February capacity review

We reviewed the “everything everwhere all the time” capacity
spreadsheet and confirmed the various people’s allocations for
February:

  • anarcat: coordination, security policy, pgBackRest, MinIO backups
  • groente: email wrap up, start work on a plan for merging
    authentication services
  • lavamind: Puppet packaging and deployments, rdsys
    contenainerization, GitLab MinIO migration
  • lelutin: Prometheus phase B, MinIO backups
  • zen: Tails’ Bitcoin retirement, LimeSurvey merge, Icinga retirement
    plan, Puppet merge plan proposal

g10k decision

we’re going to go ahead with the original g10k control repo plan (no
git modules, no monorepo, yes Puppetfile, yes git/package hashes),
this will require replacing the current environments deployment hook
provided by the puppet module and investigating how to deploy the
environments with g10k directly.

Next meeting

March 3rd, as per regular scheduling.

Metrics of the month

  • hosts in Puppet: 90, LDAP: 90, Prometheus exporters: 584
  • number of Apache servers monitored: 33, hits per second: 609
  • number of self-hosted nameservers: 6, mail servers: 90
  • pending upgrades: 0, reboots: 84
  • average load: 1.17, memory available: 3.26 TiB/5.11 TiB, running processes: 238
  • disk free/total: 58.89 TiB/142.92 TiB
  • bytes sent: 475.80 MB/s, received: 304.62 MB/s
  • GitLab tickets: 257 tickets including…
    • open: 1
    • icebox: 156
    • needs information: 4
    • backlog: 21
    • next: 16
    • doing: 6
    • needs review: 11
    • (closed: 3919)

We do not have an upgrade prediction graph as there are no major upgrades in progress.

1 Like