tmiller1 Absent Member.
Absent Member.
1350 views

Slow cluster resource load on OES/LinuxBookmark

We have a mixed node Novell cluster, with OES2 and NetWare nodes. It is mixed for the purpose of migrating to OES.

When a cluster volume resource fails over from one NetWare node to another, it is very fast, and only drops a single packet. However, when failing from one OES node to another, or from a NetWare node to an OES node, it is much slower, with three to five packets dropped. It takes 2.5 to 3 times as long to load the s

Any ideas on how to improve the performance of the OES nodes?
Labels (1)
Tags (1)
0 Likes
2 Replies
utman Absent Member.
Absent Member.

Re: Slow cluster resource load on OES/LinuxBookmark

I am not sure how all of the pieces work but I do know that since it is a mixed cluster the OES piece has to read the scripts from eDirectory so it has to make and LDAP call. You could try putting a replica of the cluster on the OES node to see if it speeds it up. The other thing is to finish your migration to OES so that the script can be converted once and for all instead of converting it each time.
0 Likes
tmiller1 Absent Member.
Absent Member.

Re: Slow cluster resource load on OES/LinuxBookmark

Putting a replica on the OES server doesn't help.

The note about converting the cluster load & unload scripts sounds reasonable,and indeed I thought of this... Do you know of any documentation that states that the resource load will be longer until the scripts are converted?

The problem that I am having is that we have a number of databases that are accessing data on these cluster resources. These databases are very sensitive to any interruption in communications. They work fine moving from NetWare node to NetWare node, but they are guaranteed to lose their connections with the extended time that it will take to move them to OES2 nodes.

My understanding is that once you run the cluster convert command, there is no rolling back the resources to be hosted on a NetWare node (and in fact there would no longer be a NetWare node in the cluster). Sooo.... I have no real fall-back plan (other than creating a new cluster, changing login scripts, etc. - very ugly!) if the performance DOESN'T improve.

I need to have some kind of documentation to cover my back end so to speak...
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.