DNS Cluster not syncing after update

[johnny]

Hello after the latest update, My DNS Records do not sync to my dns server. both servers were updated. I have rebuilt DNS and manually pushed it out using /usr/local/vesta/bin/v-update-sys-queue dns-cluster.

I see the following errors

2016-07-03 17:00:01 v-add-remote-dns-domain 'admin' 'domain.com' 'yes' [Error 4]
2016-07-03 17:00:01 v-add-remote-dns-record 'admin' 'domain.com' '28' [Error 4]

I do not understand what Error 4 means

[tjebbeke]

Read the documentation. ;)
http://vestacp.com/docs/api/#return_codes

Error 4: E_EXISTS / Object already exists
The domain and record you try to add/sync already exist on the DNS slave.

[kwanann]

Seems to be an issue, although the command appears in the error log with [Error 4], if you were to extract the command and run it by itself, it will run without any issues

Looking at the v-update-sys-queue to see if i can figure the issue..

*UPDATE
Not sure what is the exact cause, but here is what i found out

1. Run the following command, for some strange reason, dns-cluster.pipe is not marked as executable
chmod 750 /usr/local/vesta/data/queue/dns-cluster.pipe

2. Manually run the following command to update the dns-cluster
/usr/local/vesta/data/queue/dns-cluster.pipe to update the dns-cluster

[alexcy]

I have the same problem. Did you find a solution other than running manually the dns-cluster.pipe command?

Thanks.

[alexcy]

Honestly, I used to love Vesta but now I am too tired with all these bugs. I don't know if its worth it anymore.

[alexcy]

Believe me, it's not a minor bug to have 25 websites offline for a few hours because DNS were not updated! But you are right, I shouldn't have reacted that way but I was extremely frustrated and thus I apologize.

What annoys me the most is that there are bugfixes ready and the developers don't push an update! I really can't understand why do we have to wait for 8-10 months to get bugfixes via the official channels (and not github which it was many times indicated by the developers that it's not a good way to keep our systems updated).

I agree that it is the best panel. At least, it used to be until v15. Of course I have thanked the developers numerous time. I have been using VestaCP (and active in this community) for more than two years. If I am not mistaken you have joined us a few months ago?

To sum up, I should have reacted this way and I apologize.

[ScIT]

Believe me, it's not a minor bug to have 25 websites offline for a few hours because DNS were not updated! But you are right, I shouldn't have reacted that way but I was extremely frustrated and thus I apologize.

What annoys me the most is that there are bugfixes ready and the developers don't push an update! I really can't understand why do we have to wait for 8-10 months to get bugfixes via the official channels (and not github which it was many times indicated by the developers that it's not a good way to keep our systems updated).

I agree that it is the best panel. At least, it used to be until v15. Of course I have thanked the developers numerous time. I have been using VestaCP (and active in this community) for more than two years. If I am not mistaken you have joined us a few months ago?

To sum up, I should have reacted this way and I apologize.

Hi alexcy

I can understand your frustration, we have the same problems on our dns cluster. Just tried to search the error but for that, I does not understand the code as well as needed. Just get the error "Error: another sync process already running" if I try to run the code manually with enabled "output".

I hope there will be a fix soon, bad situation!

Best Regards
Raphael

[ScIT]

Hi XoXiLhJ0mn

Thanks for your answer, as requested I've added a seperate thread with my sync problem: http://forum.vestacp.com/viewtopic.php?f=13&t=12144

There is also some data with an extension pipe that locks a parallel process. I saw in this forum that if this files was not deleted after the syncing was done, then the next process would not know that the earlier job is done. In such a case, you need to delete the xxx.pipe file. Very simple.

Already tried, but thanks for the input.

It is much easier to write a small shell script to have domains rebuild the DNS files. If you transport the user data from Vesta dir to the remote server and execute the rebuild there locally, then all the DNS files will be synchronized locally rather than remotely.
Alternatively, one could make a backup and import it on the remote servers.

I would like to prefer the already existing "integrated" way for sync, this worked well since I use VestaCP.

Lets have a look, if there will be some input or ideas to solve this problem in my thread.

Best Regards
Raphael

[skurudo]

Bug on tracker
https://bugs.vestacp.com/issues/339
https://bugs.vestacp.com/issues/323