Breakpad/Status Meetings/2016-08-10

« previous meeting — index – next week » ^create?

Meeting Info

Breakpad status meetings occur on Wed at 10:00am Pacific Time.

Conference numbers:

   Vidyo: Stability 
   650-903-0800 x92 conf 98200#
   800-707-2533 (pin 369) conf 98200# 

IRC backchannel: #breakpad
Mountain View: Dancing Baby (3rd floor)

Operations Updates

P1 Infra Bugs

• python upgrade -- what is the state?
  • upgrade was underway and then it stopped
  • peterbe confirms the stage webheads are not using the new expected version of python
  • need to talk to JP about the state of things
• pingdom?
  • still waiting on accounts
• middleware outage
  • report index errors
  • phrawzty, jp, peterbe tried debugging
  • no python errors, it was happening pre-python
  • conclusion is -- bummer? ephemeral error
  • we implemented health check monitoring to catch this
  • best response is going to be getting rid of the middleware
    • this will let us shut down (some) uWSGI

Project Updates

• We have MOTD now (Message Of The Day)
  • in prod

Deployment Triage

• https://whatsdeployed.io/s-7M7

PR Triage

• http://prs.mozilla.io/mozilla/socorro

Major Projects

Migrating off of persona

• In production! Yay!
•  :njn struggles to sign in.
  • Only person to contact me with any problems.
  • Still debugging it. Apparently works on some clear Firefox profiles.
• google auth is configured to .com
  • need to add .org (DONE)

Sending public data to parquet for reading from spark/re:dash

• The CrashStorage class that sends a refined crash to S3 is in master!
  • https://github.com/mozilla/socorro/pull/3412
  • Adrian to configure it up on stage
• Scala code in progress.
  • Wrote a script that takes a JSON Schema file and generates the Scala structs source code https://gist.github.com/peterbe/9ab091bdaaebb832c09007a489de5217
• Found some crash report keys that are too ambiguous and needs to solidify before finishing the Scala code entirely https://bugzilla.mozilla.org/show_bug.cgi?id=1294088

Symbols service refactoring (snappy, somewhat tangental to us)

• ddurst will own it entirely
  • his team plans to do a wholecloth rewrite
• we can stop tracking this now

Signature generation across crash reporters

• ddurst wants this service for crash pings
  • we cannot deliver an exact match
  • we may be able to do good enough
  • https://github.com/adngdb/crash-signature-service
    • README has instructions to try that service out

Splitting out collector

On hold.

Will is spending time documenting the current collector and processor. The docs are in progress and temporarily located here:

• https://github.com/willkg/socorro-zero/blob/master/docs/arch_collector.rst
• https://github.com/willkg/socorro-zero/blob/master/docs/arch_processor.rst

Collecting client-side JavaScript errors

• was on hold while e10s shipped.
• waiting on platform to resource it

Handling more PII data in crashes

• no update

Sending stacks for all crashes from the client

• ddurst is moving ahead
  • symbol api rewrite is blocking this
  • signature generation as a service is blocking this
  • it's still an open question if that's the way we're going to go

Replacing FTPscraper

• Nick Thomas in the midst of working on generating a "socorro.json" file in the build artifacts in TaskCluster.
  • Even when we have a taskcluster scraper/crawler we're going to need to keep ftpscraper.py for Thunderbird and Seamonkey.

other business

WORK WEEEEK

• book travel this week or next
• rust conf at the tail end, buy tickets and expense if you want
  • trainings or conference

Travel, etc

• mbrandt - services qa workweek in MV, August 15 - 19

Links