chore: bump version to v1.20.0-post

[mathiasbynens]

No description provided.

[mathiasbynens]

One of the CI jobs is failing now:

Ran 676 (ok - 665, skipped - 11) of 687 test(s)
Finished in 102.015 seconds
## CRASHED ##
Message:
  UNHANDLED PROMISE REJECTION
Stack:
  Error: Page crashed!
      at Page._onTargetCrashed (C:\projects\puppeteer\node6\lib\Page.js:319:24)
      at CDPSession.Page.client.on.event (C:\projects\puppeteer\node6\lib\Page.js:149:56)
      at emitOne (events.js:96:13)
      at CDPSession.emit (events.js:188:7)
      at CDPSession._onMessage (C:\projects\puppeteer\node6\lib\Connection.js:252:12)
      at Connection.<anonymous> (C:\projects\puppeteer\node6\lib\Connection.js:138:17)
      at next (native)
      at step (C:\projects\puppeteer\node6\lib\Connection.js:99:24)
      at Promise (C:\projects\puppeteer\node6\lib\Connection.js:117:12)
      at fn (C:\projects\puppeteer\node6\lib\Connection.js:95:10)
WORKERS STATE
  0: [OK] Chromium Tracing should throw if tracing on two pages (tracing.spec.js:157:5)
=== UPLOADING Flakiness Dashboard
  > Dashboard URL: https://github.com/aslushnikov/puppeteer-flakiness-dashboard.git

Any ideas, @aslushnikov?

[aslushnikov]

@mathiasbynens this is one of the recurring problems we see on windows. There's a https://flaky.aslushnikov.com - a flakiness dashboard for all the CIs we use here. There you can see that appveyor has this recurring for quite some time.

P.S. Also, I see that PRs are now picked up by Flakiness Dashboard as well. Do you guys change anything in the appveyour and travis settings? It looks like the FLAKINESS_DASHBOARD_PASSWORD variable is no longer marked as "secure" or "encrypted". Could it be the case?

[mathiasbynens]

Do you guys change anything in the appveyour and travis settings?

I did not! I would not make changes before discussing them with you.

It looks like the FLAKINESS_DASHBOARD_PASSWORD variable is no longer marked as "secure" or "encrypted". Could it be the case?

That sounds potentially bad...