Over the last week, I upgraded most of my OES 2015 SP1 servers to OES 2018 without any real problems. This evening, I had a chance to upgrade another server to OES 2018, but after doing so, I found that none of my workstations could find their eDirectory tree. This server that I just tried upgrading was our last SLP DA running on OES 2015 SP1.
After logging into a workstation locally and running slpinfo, I see that the client thinks that it is its own DA. I remember that this was a bug when OES 2018 first came out, and I purposefully held off of upgrading any servers because of it. I wasn't really thinking about it anymore at the point when I started upgrading servers last week, and just assumed that it would have been fixed in one of the updates that have been released in the mean time. It looks like that isn't the case though.
Luckily, I have a fairly small network, without many users on the weekend, so I was able to briefly shutdown all of my eDirectory replica servers and take a VMware snapshot of each before upgrading this server, so "downgrading" the server back to OES 2015 SP1 wasn't too difficult or time consuming, but it seems that I'm going to have to hold off on upgrading for the time being.
The link that I had originally saved to a TID regarding the issue doesn't appear to be valid any more:
I see that the issue appears to be fixed in SLES 12 SP3, but not in SLES 12 SP2 LTSS, which feeds into the OES 2018 update channel.
Does anyone here have any inside knowledge as to when new OpenSLP packages might find their way into the update channel, or is this something that is unlikely to be fixed until OES 2018 SP1 is released, which would presumably be based on SLES 12 SP3, and receive the fix that way? I think that I might be stuck leaving this server on OES 2015 SP1 and running SLP without redundancy in the mean time.