Highlighted
bkuhlmann Respected Contributor.
Respected Contributor.
217 views

stuck replica error -672

Hi,

we got a stuck replica in our eDir environment.

Our eDir has 4 Replicas, the root replica is not affected.

The issue occured while deleting replicas from one of the servers in the replica ring.

At the same time one of the servers (still NetWare, subordinate replica) was rebooted.

There are 16 replicas servers for this replica mostly OES2018SP1, OES2015 and 1 NetWare.

The server after that had no replicas if you did ndsrepair on it, but the other servers in the replica ring did not get this information and said -672 at ndsrepair -E

So we got -672 errors.

Then we did a shutdown -h on the server in question.

Now we have -626 and -625 errors.

I tried TID7003130 and it worked all right but nothing happens.

So I assume the operation to remove the server from the replica ring waits for the "delete replica from server" action to finish before starting?

So I need help to get out of this situation properly and not to make things worse with my actions.

Would it be a good idea to delete the server objects from the tree?

We wanted to do this as next step to reinstall the server anyway.

 

Kind regards

Bettina

Labels (1)
0 Likes
4 Replies
bkuhlmann Respected Contributor.
Respected Contributor.

Re: stuck replica error -672

Hi,

I opened a SR in the meantime and deleted the server object following the advice of MF support.

We will see.

0 Likes
Knowledge Partner
Knowledge Partner

Re: stuck replica error -672

...which was an aggressive yet valid approach. Basically what WON'T work is downing the box while the server object still exists as even with cleaned up replica rings you'd likely face lots of obits summing up (due to xrefs). Choices are

- deleting the ncp server object along with downing the box

- keeping the ncp server object in the tree and the server up

 

0 Likes
bkuhlmann Respected Contributor.
Respected Contributor.

Re: stuck replica error -672

Hi Mathias,

yes, I thought that the "delete server from replica" operation did not work because eDir was stuck in the operation started before. Which was to remove all replicas from the server.

Well, deleting the server object has done the trick - everything behaves again.

We still do not know why the operation got stuck first with -672 and we do have to reinstall about 5 replica server till we are done and on OES2018SP1 with all of them.

Regards

Bettina

0 Likes
Knowledge Partner
Knowledge Partner

Re: stuck replica error -672

There are tons of possible reasons for a 672 (which stand for a merely generic "no access"). Pretty often, especially on removing a server from a ring with lots of replicas, it's only a temporary issue which will vanish on its own. What tends to happen in such cases (especially in mixed rings) is that a box which doesn't know yet about the fact that one player has left the field (or doesn't have it fully processed) still sends updates which would trigger a 672 on the receiving side. But that's just one of many possible reasons.

Could you please furtherelaborate on "and we do have to reinstall about 5 replica server till we are done"?

 

 

0 Likes
The opinions expressed above are the personal opinions of the authors, not of Micro Focus. By using this site, you accept the Terms of Use and Rules of Participation. Certain versions of content ("Material") accessible here may contain branding from Hewlett-Packard Company (now HP Inc.) and Hewlett Packard Enterprise Company. As of September 1, 2017, the Material is now offered by Micro Focus, a separately owned and operated company. Any reference to the HP and Hewlett Packard Enterprise/HPE marks is historical in nature, and the HP and Hewlett Packard Enterprise/HPE marks are the property of their respective owners.