teuthology/misc: Add timeout parameter to stop_daemons_of_type for better flexibility

[NitzanMordhai]

Updated stop_daemons_of_type to accept a timeout parameter, allowing dynamic control over the timeout value passed to the stop function of each daemon.

[kshtsk]

Is this tested already?

[NitzanMordhai]

Is this tested already?

i ran few test using my PR issue. didn't riched to timeout, but the param was passed

[VallariAg]

@NitzanMordhai the timeout didn't work even though the param was passed?
I'm not sure if timeouts are actually used in all kinds of daemon's stop() method:
It's used here:

teuthology/teuthology/orchestra/daemon/state.py

Line 131 in cca92a2

def stop(self, timeout=300):

But I don't see it being used in these overrides of systemd and cephadm daemons:

teuthology/teuthology/orchestra/daemon/systemd.py

Line 199 in cca92a2

def stop(self, timeout=300):

teuthology/teuthology/orchestra/daemon/cephadmunit.py

Line 154 in cca92a2

def stop(self, timeout=300):

[NitzanMordhai]

@NitzanMordhai the timeout didn't work even though the param was passed?
I'm not sure if timeouts are actually used in all kinds of daemon's stop() method:
It's used here:

that's the purpose of that PR, to add the timeout so we can use it and not use the default

[VallariAg]

@NitzanMordhai This PR relies on passing timeout to daemon.stop(timeout). But I'm not sure if this stop method actually uses the timeout param (depending on daemon type).

[NitzanMordhai]

it will be pass to teuthology/orchestra/daemon/state.py
that will wait until the daemon stops (or timeout)

[VallariAg]

Great, it should work well then!

[rzarzynski]

@VallariAg, @NitzanMordhai: is this ready to merge? If it is, let's not hesitate and unblock the PR in ceph.