Sign up on the Revelation Software website to have access to the most current content, and to be able to ask questions and get answers from the Revelation community

At 11 NOV 2002 11:00:18AM Oystein Reigem wrote:

Can Flush and GarbageCollect go wrong? Should one use Set_Status and Get_Status to catch possible errors?

- Oystein -


At 11 NOV 2002 01:07PM Richard Hunt wrote:

Oystein,

When a "GarbageCollect" occurs, all it does is "defrag" the OI memory. That way the available memory is in a bigger chunk and not in small chunks. I doubt it would speed things up any.

"GarbageCollect" is automatic. Once AREV or OI requires more memory, it checks for available memory. If the available memory does not meet the requirements then a "GarbageCollect" happens and available memory is then checked again.

"GarbageCollect"ing manually probably would be not very useful, and rather is probably very useless.

If a "GarbageCollect" fails… well I would think its kinda all over! And a crash is about to happen, due to corupted memory.

Unless you are attempting to prevent exceeding memory limits before you exceed memory limits, I would not worry about doing a "GarbageCollect" since it is done automatically when it is needed.

Now "flush" is much different. Since it forces completion of any buffered and pending writes, it is a necessity sometimes. Although I do not use it, some advanced programmers might. If a "flush" fails, then I would guess that the advanced programmer would want to know. And the more I think about "flush" the more I believe that a "flush" failure would probably crash OI also. So I would guess that knowing in advance that a "flush" failed might just be useless. Maybe?

Just my opinion.


At 12 NOV 2002 04:33AM Oystein Reigem wrote:

Richard,

Thanks. Somebody recommended doing Flush and GarbageCollect before detaching and attaching volumes. That's my interest in Flush and GarbageCollect.

- Oystein -


At 12 NOV 2002 05:32AM [url=http://www.sprezzatura.com" onMouseOver=window.status= Click here to visit our web site?';return(true)]The Sprezzatura Group[/url] wrote:

Oy

Theoretically FLUSH should clear out the FRAMES array (see REVMEDIA passim and call each BFS/MFS telling it to flush. This in turn SHOULD set @File.Error if there was a problem but in honesty it is difficult to conceive of such except with not linear hash BFSs.

The Sprezzatura Group

World Leaders in all things RevSoft


At 12 NOV 2002 06:42AM Oystein Reigem wrote:

Sprezz,

Thanks.

Assume if I just had put in my Get_ and Set_Status calls there would be no harmful consequences and less energy expended for all of us. But I couldn't resist. :-)

- Oystein -

Uncontrollable urge

Got an urge got a surge

And it's out of control

Got an urge I wanna purge

'Cause I'm losing control

Uncontrollable urge

I wanna tell you all about it

Uncontrollable urge

Makes me scream and shout it

It's got style it's got class

So strong I can't let it pass

I gotta tell you all about it

I gotta scream and shout it

And I say yeah

Devo, 1978


At 13 NOV 2002 04:26AM Richard Bright wrote:

The someone was Cam Purdy.

In earlier versions prior to OI 3.7 a had serious problems when swapping volumes. The Flush / garbagecollect process seemed to resolve this - essentually clearing out cached file handles.

Cant say it is required - bit like other forms of supistitious behaviour.

'cause we all bow to Sprez wisdom.

Richard Bright

View this thread on the forum...

  • third_party_content/community/commentary/forums_nonworks/10ae49be50ac279085256c6e0057eb65.txt
  • Last modified: 2023/12/28 07:40
  • by 127.0.0.1