[PATCH] smbcontrol test fix

[Douglas Bagnall]

Andrew wrote:
> We came up with a plan for smbcontrol in the office yesterday, so that
> we should get fixed in coming days

The test is trying to ensure that the processes known to
irpc_all_servers() are pingable, and the problem is that there are
fleeting ldap_server processes that disappear before the ping.

What we do here is check the processes again when one fails to answer.
If it has disappeared from the list then either it was one of those
transitory processes and we can relax, or something very serious is
going on (like perhaps our ping killed it). We attempt to distinguish
between those cases by checking whether there is still a server that
shares the dead one's name, as we'd expect in the forking case. I know
you can imagine scenarios that escape this check, but I suspect they
can be categorised under at least one of "we're hosed anyway",
"outside the scope of this test", and "do you want this to keep
flapping forever?".

cheers,
Douglas
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-tests-smbcontrol-reduce-ping-test-false-positive-rat.patch
Type: text/x-patch
Size: 3535 bytes
Desc: not available
URL: <http://lists.samba.org/pipermail/samba-technical/attachments/20180304/47d08e50/0001-tests-smbcontrol-reduce-ping-test-false-positive-rat.bin>