Multicast is Slow
-
I have a ubuntu 14.04.1 server running fog 3658 that multicast is running slow on. I am getting about 30MB/min. Our old fog server running .32 never had any issues with multicast and the network setup is the same so I don’t think the issue is there, but I’m not positive.
-
You should update to something like 3780 when sourceforge comes back up completely. Around that version, @Tom-Elliott added a throttle for multicast that you can set.
The revision you’re using is - round about the time - when Tom was tinkering with the multicast stuff.
you could also try the latest FOG Trunk, too.
But as with any update - always always export your DB and hosts before hand.
-
Thanks! I’ll update it as soon as I can and post back here whether or not it is fixed.
-
After updating, I am still having issues with multicast. Sometimes multicast will just freeze on starting to restore image. It used to do that before the update, but after restarting a few times it would work, but it would be unbearably slow. I just checked the multicast log and it says “This is not the master node.” I don’t remember seeing that message before the update. I only have one node set. I checked the interface the storage node is using and changed that to em1 to see if that would fix it and it didn’t. I am also having issues with unicast now where it hangs on checking in. Any help is appreciated.
-
@ryanxplosion Not trying to ignore or side-skirt your multicast issue, but due to the complexity of Multicast problems and your need to get work done, I would advise you to begin using Unicast to get your workload done while you continue to search for an answer to multicast. Don’t let this problem stop you from getting work done.
-
I ended up restoring from a backup I did last week before the update and ran the update again. I didn’t have any issues this time and multicast is working properly again! Thank you!
-
I may have spoke too soon. It worked once and now it just says starting to restore image (-) to device (/dev/sda1). The log is saying this is not the master node. Any idea what I should look at for that?
-
@ryanxplosion Each storage group can have one master node, and no more than one master node.
If you only have one group and only one node within the group, then you can safely set that node to master. See if the error persists.
-
@Wayne-Workman I just have the default storage group and master node is checked in the GUI for that group.
-
@ryanxplosion Then uncheck it, save, wait a moment, then re-check it and save.
-
@Wayne-Workman Same issue
-
@ryanxplosion Make sure the image is assigned to the correct storage group.
-
@Wayne-Workman I just double checked and the image is part of that storage group. Any other suggestions?
-
@ryanxplosion Well… You said it worked once… What happens if you reboot and try?
-
@Wayne-Workman Rebooting didn’t help. I ended up running the installer again and that fixed it. I had some issues with the tasks not ending properly after it was completed and the service would get stuck. I just cleared the multicast tasks in mysql and restarted the multicast service and that seemed to work. Haven’t had any more issues since that happened earlier. Thanks again for all of your help.