Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations SkipVought on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Process Error 153 - Enterprise Server 9.5 SP1

Status
Not open for further replies.

Stelex

IS-IT--Management
Dec 25, 2006
15
0
0
I recently upgraded from 9.2 to 9.5 SP1 using the clean install method (Clean instal of 9.5 SP1 with an exact copy of the 9.2 Database). Everything went well until I tried to start the Index Server to recreate the index.

I an now getting the following error: The Process has returned: Process Error 153: The Update Distributor process failed to start because one or more of the parameters in the search.ini file are invalid"

Anyone have a quick and easy way of solving this problem? I've already deleted the OTAdmin files and the search.ini file, rebooted the server, etc to no avail.
 
If you are sure that you removed otadmin.cfg its old
and otadmin.cfg ,its old and serach.ini maybe it is time to call ot.also make sure on your default admin server ,rmi port has information.that is the first I would say to check

Well, if I called the wrong number, why did you answer the phone?
James Thurber, New Yorker cartoon caption, June 5, 1937
 
I think the RMI port is missing from search.ini.
I've seen this before.

Also check if the DCS parameters are set up correct in the dataflow. I've seen cases like this where the DCS parameters were empty after an upgrade.
 
Did you check the kini table in your database. It might be having invalid entries. In the clean install method did you clean up kini table entries (not all, but specific rows).

Opening support ticket will be useful on this.
 
Actually - the solution was obvious - we were looking for a forest and all we saw were trees.

When we attempted to recreate the index after initially attaching the database, we had 2 options - Resynchronize and Continue - unfortunately we picked Resynchronize (even though the migration script specifically said "Continue").

By the time we realized the error - the DTree Table and KIni was so messed up we had to go in and manually delete the references to the OLD server using the following SQL Queries:

1. On the System Object Volume page, we checked the Slice Folder to see if the "Enterprise" and "Enterprise [All Versions]" slices were present. We deleted the references to the Enterprise out of the slice folder, and then checked to see if "Enterprise Data Source" was in the Add Item list. It wasn't (sigh)

2. We then executed the following SQL on the Livelink db:

select * from kini where inisection='SliceReferences' and inikeyword like 'Library%'

We then deleted all of the rows returned. That didn't work either.

3. Finally, we ran the following query:

select DataID, SubType, Name from DTree where Name like 'Enterprise%'

This gave us a list of everything that had a name that began with "Enterprise". We deleted everything except the Enterprise Workspace (2000)

To finally clean things up we ran a couple extra queries as follows:

delete from kini where inisection='SliceReferences' and inikeyword like 'Library%'

delete from kini where inisection='UniqueSliceReference' and inikeyword like 'Enterprise%'

Looks like a lot of work - but at least we now had an indexable Enterprise Workspace.
 
This should have worked just fine.I just did this on a copy of a prodn snapshot.This is probably the 15 th time I am creating a duplicate of prodn as test without actually messing up...
Livelink services and Admin should be down
a)Delete from kini where inisection ='AdminServers'
This ensures all previous references of the prod server is gone and you will not mess up prodn.
b)remove the four files as in OT document and search.ini if you have it
c)Restart Admin,Llserver It will pick up the server you are on.For added safety try to do this locally on the box that will be your admin.
d)Now check the RMI port by clicking properties-general of your default admin server.It will be missing the ports.Type two ports and the log file and update it.This is all that is necessary for making the Admin server do its work.
e)Now you can delete the 'Enterprise Data Source' folder.
f)Now change the URL to ?func=ipool.autocreation the search partition wizard starts up and lo and behold everything
will work again....


One thing you want to be really concerned about making a dev system out of a prod snapshot is livelink 9.5/9.5SP1 has a bug so it does not have a GUI command to change the External file store.You have to manipulate KINI manually on these versions even though making the update statement itself is a work of art.....

Just wrote this because i used to painstatkingly do those kini things in the past but this would save you some time.

Well, if I called the wrong number, why did you answer the phone?
James Thurber, New Yorker cartoon caption, June 5, 1937
 
Also I forgot to mention thi si in the KB creating an enterprise datasource if it did not get created automatically.I just used the prod>dEv+ this piece effectively.Seems to be the most less trouble some method for replication in my opinion.

Well, if I called the wrong number, why did you answer the phone?
James Thurber, New Yorker cartoon caption, June 5, 1937
 
Yeah - unfortunately ours was so messed up I couldn't delete the references from the KIni file the way you described, so I had to go the hard way.

I didn't know the ipool.autocreation url - thanks for that tip!
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top