Avi Kivity

``` Results (609.51s (0:10:09)): 92 passed 8 failed - cluster/test_tablets.py:1310 test_two_tablets_concurrent_repair_and_migration_repair_writer_level[3-100].debug.1 - cluster/test_tablets.py:1310 test_two_tablets_concurrent_repair_and_migration_repair_writer_level[30-100].debug.1 - cluster/test_tablets.py:1310 test_two_tablets_concurrent_repair_and_migration_repair_writer_level[46-100].debug.1 - cluster/test_tablets.py:1310 test_two_tablets_concurrent_repair_and_migration_repair_writer_level[44-100].debug.1 - cluster/test_tablets.py:1310 test_two_tablets_concurrent_repair_and_migration_repair_writer_level[66-100].debug.1 - cluster/test_tablets.py:1310 test_two_tablets_concurrent_repair_and_migration_repair_writer_level[94-100].debug.1 - cluster/test_tablets.py:1310 test_two_tablets_concurrent_repair_and_migration_repair_writer_level[100-100].debug.1 -...

> This is the condition variable which times out: > > ``` > 561 return utils::get_local_injector().inject("repair_writer_impl_create_writer_wait", utils::wait_for_message(200s)); > ``` > > Looks like 200s is not enough anymore. 200s is...

Yes, when the test is successful, it completes in a few seconds.

Bisect is still running, but converging on faad0167d7a1c26aab73e37bb74b69ec5e77d837

\# bad: [8822c23ad4ee6fc783862671a3a4472dfc6778b5] Merge 'test: cqlpy: test_protocol_exceptions.py: increase cpp exceptions thr…' from Dario Mirovic \# good: [4784e39665c0aea5a2ab2c3c53ca181752bdba75] auth: fix ctor signature of certificate_authenticator git bisect start '8822c23ad4ee6fc783862671a3a4472dfc6778b5' '4784e39665c0aea5a2ab2c3c53ca181752bdba75' \# bad:...

https://github.com/scylladb/scylladb/pull/27560

Looks like chunked_fifo wins. And again I need to re-measure https://github.com/avikivity/seastar/commits/sort-tasks/, I found some bugs in it.

My feeling is that this thing is more sensitive to different workloads than chunked_fifo. A run with very small+fast tasks will prefer large F, so that prefetch can happen in...

> > Would it be worth considering chunked_fifo as a first iteration? > > It's better to discuss it in #3125 , because that's the place where the new collection...

What about annotations of the CI failures? If we don't track them, they will never be fixed.