mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Joseph Wu" <jos...@mesosphere.io>
Subject Re: Review Request 39365: HTTP Scheduler should abort when MasterDetector create fails.
Date Fri, 16 Oct 2015 00:26:51 GMT


> On Oct. 15, 2015, 5:13 p.m., Vinod Kone wrote:
> > src/scheduler/scheduler.cpp, line 170
> > <https://reviews.apache.org/r/39365/diff/2/?file=1099498#file1099498line170>
> >
> >     For posterity, can you add a blurb in the review on what the exact issue is?
> >     
> >     Also, are the other error() calls in this method safe to execute?

Can do.

There are 2 other `error()` calls:
1) In the initialization, in case the flags fail to parse.  This one will lead to a segfault.
2) In the master detection loop (after initialization).  This one is ok, since we handle it
by retrying in a loop.


- Joseph


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/39365/#review102841
-----------------------------------------------------------


On Oct. 15, 2015, 5:03 p.m., Joseph Wu wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/39365/
> -----------------------------------------------------------
> 
> (Updated Oct. 15, 2015, 5:03 p.m.)
> 
> 
> Review request for mesos, Anand Mazumdar and Artem Harutyunyan.
> 
> 
> Bugs: MESOS-3748
>     https://issues.apache.org/jira/browse/MESOS-3748
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> If not, the process will segfault during initialization.
> 
> 
> Diffs
> -----
> 
>   src/scheduler/scheduler.cpp 56801ca6ffc9f9f0e4bd12dbf535e9c5251c2712 
> 
> Diff: https://reviews.apache.org/r/39365/diff/
> 
> 
> Testing
> -------
> 
> `make check`
> 
> ```
> export DEFAULT_PRINCIPAL=root
> build/src/event-call-framework --master="asdf://127.0.0.1:5050"
> ```
> Check that it does not segfault.
> 
> 
> Thanks,
> 
> Joseph Wu
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message