Search results

There are many reason for locks hanging around a long time - and imho they are almost always due to poor programming in one way or another. I fairly often find that it's because of a missing release. Some programmers think that at the end of a transaction all locks "go away".

Yes Rob, I am aware of that. I actually take a bit more just because of that. I am only interested in troublesome locks, in particular those that have been around "a bit too long" so I can identify programs that are holding locks too long - most likely due to the omission of a RELEASE command...

Tom - you are quite correct (of course!) - promon will continue providing as many pages as you ask for. Therefore, one can use _DbStatus._DbStatus-NumLocks to determine how many pages to request. Thanks everyone - I believe my problem is now solved.

Does anyone know what promon does if there are more than 9999 locks? I expect that it will just give you the first 9999 - and that's it. Also - is anyone aware of any plans from Progress to increase the number of locks promon can report?

That's my experience too. On a "big" system under heavy load - ONE scan of the lock table with _Lock can take more than one hour (!!) - and at the same time it creates a terrible impact of system performance. That's exactly why I've moved to using promon. It provide a full copy of everything...

I would think that gives you the size of the lock table - not the number of locks actually in it.

How do you check how many locks are in the table before scanning it?

Thanks Tom. I was pretty sure that was the case - but I thought I'd better ask .... just to be sure. (Pity about the 9999 limit.)

Hi, OE 10.2B - RH Linux. I have a process that gets database lock details using the output from promon. I used to do it with the _Lock VST, but found that too slow. Recently an associate said there was a risk of "freezing" a database by using promon to do this on "large" databases. That...

Rob - I agree with you entirely. What you have said is almost exactly what I said myself about the matter! But ..... deaf ears, I guess. Tom - thank you very much. I believe you have "quantified" OEE and OEM for me. I think I'll let them gather a bit more dust.

Thanks for the comments, Rob. Your point about rolling-out 10.2B is well-understood. I have pushed that barrow - but the decision is not mine to make. There are many balls in the air here right now - and the priority is to get off SCO/9.1D ASAP. The belief is that once all sites are RH/10.2B...

We have several 10.2B databases at a central office and about 250 others connected remotely. Most hosts are Linux (RH) - and some are SCO. Half the remote DBs are 9.1D - but are being migrated to 10.2B. All applications use character clients (we don't have an Appserver). I know very little...

Thank you very much for that. (But - how sad!)

I have a system that controls some overnight activity. I created a directory /u0/sr - and under it several other directories for scripts, logs, data files and a small Progress DB (10.2B under Red Hat Linux). The DB is (naturally enough) in /u0/sr/db. I ran into file space problems. It was...

This is not a DB question - but I thought the wise people in this forum may be able to help me. We have several Progress DBs serving users 24x7 in different countries. Admins are not always on-site and if a problem happens we need to alert people. This is done now with email - but that's not...

Yes - there is quite a lot of temp file activity. The "issue" about changing to 64-bit (which we physically have) is that our management will not allow the change unless the vendor has given an assurance that they have fully tested their application under 64-bit. Why doesn't the vendor...

I wasn't aware of the -Bt startup parameter; thank you for that. But I don't think that helps us. We recently upgraded our server - and we have "lots" of memory. If we "could" - we'd increase -B considerably - but we can't because we're stuck with a 32-bit version of OpenEdge (10.2B). We can't...

At a previous company where I used Solaris I put the client temp files in the /tmp directory (effectively RAM) and that gave a positive performance boost. I'm now on an AIX system (OE 10.2B on AIX 7.1) and there is a "lot" of spare memory (over 20GB). I have tested creating a RAM disk and...

I may not know the particular meaning you have in mind, Tom! But if it is oriented towards something like "awkward" ... then I sure wouldn't mount an argument against it! :p Nevertheless - there are a few things it is useful for.

I have solved the problem. Apparently an if statement is "invalid" unless it is inside a block. I enclosed everything inside {} braces - and now awk is happy. Strange limitation. Ron.