Re: [Gluster-devel] [Gluster-users] Community Meeting, Sept 26 15:00 UTC in #gluster-meeting

2018-09-25 Thread Amar Tumballi
Updated the sheet with below details, happy to hear more from our users: - - Date: 2018-09-26 - Host: Amar Tumballi - Location: #gluster-meeting on Freenode (webchat ) - Time: - 15:00 UTC - or in your local

[Gluster-devel] Community Meeting, Sept 26 15:00 UTC in #gluster-meeting

2018-09-25 Thread Amye Scavarda
Hey there! We've got a community meeting upcoming tomorrow at 15:00 UTC. This will be in #gluster-meeting on freenode, and our agenda (which is always editable!) is at https://bit.ly/gluster-community-meetings. - amye -- Amye Scavarda | a...@redhat.com | Gluster Community Lead

[Gluster-devel] Gluster performance improvements

2018-09-25 Thread Xavi Hernandez
Hi, we are starting to design the next cache implementation for gluster that should provide much better latencies, increasing performance. The document [1] with the high level approach will be used as a starting point to design the final architecture. Any comments will be highly appreciated so

Re: [Gluster-devel] [Gluster-users] Crash in glusterfs!!!

2018-09-25 Thread ABHISHEK PALIWAL
Hi Pranith, I have some questions if you can answer them: What in LIBC exit() routine has resulted in SIGSEGV in this case ? - Why the call trace always point to LIBC exit() in all these crash instances on gluster ? - Can there be any connection between LIBC exit() crash and SIGTERM handling

Re: [Gluster-devel] [Gluster-users] Crash in glusterfs!!!

2018-09-25 Thread Pranith Kumar Karampuri
On Tue, Sep 25, 2018 at 2:17 PM ABHISHEK PALIWAL wrote: > I don't have the step to reproduce, but its a race condition where it > seems cleanup_and_exit() is accessing the data structure which are not yet > initialised (as gluster is in starting phase), due to SIGTERM/SIGINT is > sent in

Re: [Gluster-devel] [Gluster-users] Crash in glusterfs!!!

2018-09-25 Thread ABHISHEK PALIWAL
I don't have the step to reproduce, but its a race condition where it seems cleanup_and_exit() is accessing the data structure which are not yet initialised (as gluster is in starting phase), due to SIGTERM/SIGINT is sent in between. Regards, Abhishek On Mon, Sep 24, 2018 at 9:11 PM Pranith