JR Richardson
2007-Apr-26 17:31 UTC
[asterisk-users] Re: Voicemail on Different Server, Voicemail with NFS
> -----Original Message----- > From: JR Richardson [mailto:jmr.richardson@gmail.com] > Sent: Saturday, June 17, 2006 2:30 PM > To: asterisk-users@lists.digium.com; Douglas Garstang > Subject: Voicemail with NFS (working, I think) > > I'm using a stand-alone VM server and exporting the VM files ro for > MWI function only. All my registration servers mount the remote NFS > share just to check MWI, all read-write functions to the VM files > occur on the VM server only. > > On the registration servers, I mounted the remote VM share with this > in my fstab.conf: > > 10.10.14.124:/var/spool/asterisk/voicemail /mnt/vmserver nfs > soft,nolock,timeo=1,retrans=1,bg,intr > > all on one line of course > > [10.10.14.124:/var/spool/asterisk/voicemail] is the voicemail server > and [/mn/vmserver] is the mounted location, [nfs] is the file system > type, > > options: (the important stuff) > [soft] If an NFS file operation has a major timeout, then report an > I/O error to the calling program. The default is to continue > retrying NFS file operations indefinitely. > [nolock] does not use locking. > [timeo=1] redces the nfs share minor timeouts to 1 tenth of a second > [retrans=1] reduces the number of minor timeouts and retransmitts > needed to cause a major timeout. > [bg] background the re-mount operation > [intr] interupt the nfs file operation for the calling program when a > major timeout occurs > > When mounting the nfs without these options, Asterisk became > immediately unstable when the remote nfs dropped off-line, but also > regained is operation immediately when the nfs became available again. > > When mouting the nfs with these options, asterisk kept on ticking > without interruption, WMI was unavailable of course, but all other > functions worked. The MWI came back on immediately after the nfs > share was restored. > > Of course this is a hack and I'm patiently waiting for Digium to > implement the Remote MWI over IAX and incorporate into stable > releases. > > Hope this helps. > > JR > > -- > JR Richardson > Engineering for the Masses
Porier, Jeremy M.
2007-Apr-27 07:37 UTC
[asterisk-users] Re: Voicemail on Different Server, Voicemail with NFS
I was afraid of an unavailable NFS mount hanging the app and I also wanted to keep all of the communication over IAX for simplicity sake. I also hacked together my own "MWI over IAX". I did write ups of how I did both. http://opensourcemadness.blogspot.com/2007/03/centralizing-asterisk-voic email.html http://opensourcemadness.blogspot.com/2007/03/propagating-asterisk-mwi-a cross.html - Jeremy -----Original Message----- From: asterisk-users-bounces@lists.digium.com [mailto:asterisk-users-bounces@lists.digium.com] On Behalf Of JR Richardson Sent: Thursday, April 26, 2007 6:32 PM To: asterisk-users@lists.digium.com Subject: [asterisk-users] Re: Voicemail on Different Server,Voicemail with NFS> -----Original Message----- > From: JR Richardson [mailto:jmr.richardson@gmail.com] > Sent: Saturday, June 17, 2006 2:30 PM > To: asterisk-users@lists.digium.com; Douglas Garstang > Subject: Voicemail with NFS (working, I think) > > I'm using a stand-alone VM server and exporting the VM files ro for > MWI function only. All my registration servers mount the remote NFS > share just to check MWI, all read-write functions to the VM files > occur on the VM server only. > > On the registration servers, I mounted the remote VM share with this > in my fstab.conf: > > 10.10.14.124:/var/spool/asterisk/voicemail /mnt/vmserver nfs > soft,nolock,timeo=1,retrans=1,bg,intr > > all on one line of course > > [10.10.14.124:/var/spool/asterisk/voicemail] is the voicemail server > and [/mn/vmserver] is the mounted location, [nfs] is the file system > type, > > options: (the important stuff) > [soft] If an NFS file operation has a major timeout, then report an > I/O error to the calling program. The default is to continue retrying> NFS file operations indefinitely. > [nolock] does not use locking. > [timeo=1] redces the nfs share minor timeouts to 1 tenth of a second > [retrans=1] reduces the number of minor timeouts and retransmitts > needed to cause a major timeout. > [bg] background the re-mount operation [intr] interupt the nfs file > operation for the calling program when a major timeout occurs > > When mounting the nfs without these options, Asterisk became > immediately unstable when the remote nfs dropped off-line, but also > regained is operation immediately when the nfs became available again. > > When mouting the nfs with these options, asterisk kept on ticking > without interruption, WMI was unavailable of course, but all other > functions worked. The MWI came back on immediately after the nfs > share was restored. > > Of course this is a hack and I'm patiently waiting for Digium to > implement the Remote MWI over IAX and incorporate into stable > releases. > > Hope this helps. > > JR > > -- > JR Richardson > Engineering for the Masses_______________________________________________ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users