[PATCH] python indent bugfix in dns_hub.py

Tim Beale timbeale at catalyst.net.nz
Tue Jan 22 23:54:21 UTC 2019


I looked more closely at the failed logs. It appears that the slow DNS
reply is not because the DC has only just started up.

E.g. in Isaac's log, the very first test case is run successfully
against the fl2000dc. It's not until ~24 minutes later that we try to
start the vampire_2000_dc and join it to fl2000dc.
https://gitlab.com/samba-team/devel/samba/-/jobs/148388249

So I don't think putting a check in wait_for_start() would help avoid
this problem.

On 23/01/19 11:33 AM, Jeremy Allison wrote:
> On Tue, Jan 22, 2019 at 09:32:43PM +0100, Stefan Metzmacher via samba-technical wrote:
>> Am 22.01.19 um 18:02 schrieb Jeremy Allison via samba-technical:
>>> On Tue, Jan 22, 2019 at 09:54:06AM +0200, Isaac Boukris via samba-technical wrote:
>>>>> The 2 second thing was nagging at me. Checking
>>>>> libcli/dns/dns.c we have:
>>>>>
>>>>> #define DNS_REQUEST_TIMEOUT 2
>>>> FYI, changing DNS_REQUEST_TIMEOUT to 10 or to 30 does not help in my
>>>> local reproduction.
>>>>
>>>> I think timeouts might sometimes be ignored when receiving a network
>>>> error such as port unreachable an so.
>>> FYI, Changing DNS_REQUEST_TIMEOUT to 5 passed gitlab-ci
>>> for me.
>>>
>>> https://gitlab.com/samba-team/devel/samba/pipelines/44257355
>>>
>>> I think a combination of this change plus adding retries
>>> in the join.py code should do the trick.
>> I'd really like to avoid retries in command line code,
>> this should just work without them. If we have problems
>> in our selftest environment, we should have retries in
>> selftest/target/Samba4.pm: wait_for_start() to make sure everything is
>> ready.
> OK, fair enough (although I already created the patches - at
> least I learned some more python :-). Do you want the command
> line tools to just ignore slow servers/timeouts rather than
> retry ? Is this a problem in the non-test outside world ?



More information about the samba-technical mailing list