Index Limit - VMS 7.3-2

[Jorge Cocomess]

Hi,

Is there such thing as Index Limit within the VMS directory or some sort? I know there is a version limit but not sure about Index Limitation. As you can tell, I am also new to VMS.

Thank you in advance.
J

[Karl Rohwedder]

Not quite sure, what you mean:
- a disk volume has a limit on how many files are allowed, specified during INIT
- when creating new files, the index file on [000000] fills up and is extended and may eventually fill up its fileheader, then no more files can be created. You may preallocate room for entries in the indexfile when INITing the disk
- too many files in one directory may slow down performance during certain operations (not a real limit...)

regards Kalle

[Jorge Cocomess]

Is there a way to raise this limit after the fact? How do I check to see how many files were set?

Thanks,
J

[Karl Rohwedder · Accepted solution]

To raise the 'maxiumum files allowed', the disk has to be reINIted, be sure to have a valid backup and do NOT forget to specify /NOINIT when restoring the volume, else the disk gets INITed by BACKUP with its original parameters.
If your problem is the indexfile, only ha BACKUP/RESTORE will help (or delete some files:-))

You may always specify a high number for /MAXIMUM_FILES, it costs nothing.

A good utility for disks is the DFU utility, often mentioned here and available via the OpenVMS freeware pages. It has a REPORT function, which tells all about your disk.

regards Kalle

[Volker Halle]

Jorge,

SHOW DEV/FULL disk will show 'Maximum files allowed'

Are you trying to solve a real problem or is this question for your educational purposes only ? If you have a real problem, it would be good to provide the full error message and circumstances, under which the problem is seen.

Volker.

[Jim Lahman_1]

How would I see the maximum number of files allocated on a disk?

- Jim

[Ian Miller.]

Jim,
DFU REPORT
should tell you number of files and max allowed.

[Jorge Cocomess]

Volker - According to my software vendor, they said, "your Index Limit" was reached and caused a certain process to hang because it couldn't create any more logs. Well, I know that I am new to the VMS world, but I thought that our "Maximum files allowed" was set at 15926636 when we initialized this disk.

Here's the output of the disk with the "Show dev/full $1$dga31:

~Many thanks,
J


Disk $1$DGA32: (ELMO), device type HSG80, is online, mounted, file-oriented
device, shareable, available to cluster, device has multiple I/O paths,
error logging is enabled.

Error count 1 Operations completed 421292290
Owner process "" Owner UIC [SYSTEM]
Owner process ID 00000000 Dev Prot S:RWPL,O:RWPL,G:R,W
Reference count 379 Default buffer size 512
Current preferred CPU Id 2 Fastpath 1
WWID 01000010:6000-1FE1-000B-B1C0-0009-1090-2086-000E
Total blocks 286679457 Sectors per track 96
Total cylinders 31107 Tracks per cylinder 96
Logical Volume Size 286679457 Expansion Size Limit 286752768
Allocation class 1

Volume label "DISK31" Relative volume number 0
Cluster size 8 Transaction count 383
Free blocks 68756800 Maximum files allowed 15926636
Extend quantity 5 Mount count 1
Mount status System Cache name "_$10$DKA600:XQPCACHE"
Extent cache size 64 Maximum blocks in extent cache 6875680
File ID cache size 64 Blocks in extent cache 153720
Quota cache size 0 Maximum buffers in FCP cache 4270
Volume owner UIC [DGBMASTER] Vol Prot S:RWCD,O:RWCD,G:RWCD,W:RWCD

Volume Status: ODS-2, subject to mount verification, write-back caching
enabled.

Device error count was last reset on: 28-JUL-2006 09:36:32.71


I/O paths to device 4
Path PGA0.5000-1FE1-000B-B1C2 (ELMO), primary path, current path.
Error count 1 Operations completed 421136684
Path PGA0.5000-1FE1-000B-B1C4 (ELMO).
Error count 0 Operations completed 51873
Path PGB0.5000-1FE1-000B-B1C1 (ELMO).
Error count 0 Operations completed 51870
Path PGB0.5000-1FE1-000B-B1C3 (ELMO).
Error count 0 Operations completed 51863

[Andy Bustamante]

It's possible your vendor means "version limit" on a file. OpenVMS allows up to 32767 versions of a file, for example: application.log;32767 is a possible file name. After this limit is reached attempts to create a new version of the file will fail. Purge won't solve the issue since it leaves the highest versions, you either need to delete or rename the file(s).

Andy

[Jan van den Ende]

Jorge,

from the info seen so far I would guess that Andy hit the point.

What is your VMS version? Recent versions have simple commands to find out if he is right; if not, then there also are ways, albeit somewhat less simple.

If we know IF he guessed right or wrong, we can proceed from there.

Please help us to help you.

Proost.

Have one on me.

jpe

[Jorge Cocomess]

Nope, it wasn't the version limit issue. I wish I knew what they purged. I will find out more tomorrow and post the result.

Thanks,
J

[Andy Bustamante]

If it's not the version limit then Karl is probably on track. The file [000000]indexf.sys contains the file headers for disk (the /headers=xxx in the disk INIT string). This file can extend on the fly, but is limited to 1 header. Once you've reached that point, any attempt to create a new file on that disk will fail. PURGE and DELETE are the short term fix. As Karl mentions above, you'll need to reinit the disk to correct the issue long term. DFU will provide information on the number of free headers.

See the help for /max and /head under INIT. The values will depend on your application. DFU will help you plan disk configuration.

Andy

[David Jones_21]

"If your problem is the indexfile, only a BACKUP/RESTORE will help (or delete some files:-))"

The Disk File Optimizer product has a DEFRAG OFFLINE_VOLUME command that will defragment the index.sys file. The disk has to be dismounted to perform the operation.

[Jorge Cocomess]

Sounds complicated. We do have license for CA product called Diskkeeper, should I be using this to defrag?

Thanks,
J

[Volker Halle]

Jorge,

did your application receive the

HEADERFULL, file header is full

error ? This would indicate, that the file header of INDEXF.SYS cannot be extended anymore.

You can easily check, whether this could be this case (if you know the disk on which this error has happened):

$ DUMP/BLOCK=COUNT=0/HEADER disk:[000000]INDEXF.SYS

If you see lots and lots of Retrieval Pointers in the Map Area (at the bottom of the file header), chances are that the INDEXF.SYS file header is full and cannot be extended anymore to allow for more files to be created on that disk.

See ATW article 4452:

http://h71000.www7.hp.com/wizard/wiz_4452.html

If this is the case and Diskeeper can defrag INDEXF.SYS, you should run it and have it just try to defragment INDEXF.SYS. You will probably need to dismount the volume temporarily to defrag INDEXF.SYS.

The HP OpenVMS Disk File Optimizer (DFO) is described at:

http://h71000.www7.hp.com/openvms/storage/dfofaq.html

Volker.

[Jan van den Ende]

Jorge,

from your Forum Profile:


I have assigned points to 176 of 278 responses to my questions.

Your profile even has TWO pages with pointers to questions with unassigned answers.

Maybe you can find some time to do some assigning?

http://forums1.itrc.hp.com/service/forums/helptips.do?#33

Mind, I do NOT say you necessarily need to give lots of points. It is fully up to _YOU_ to decide how many. If you consider an answer is not deserving any points, you can also assign 0 ( = zero ) points, and then that answer will no longer be counted as unassigned.
Consider, that every poster took at least the trouble of posting for you!

To easily find your streams with unassigned points, click your own name somewhere.
This will bring up your profile.
Near the bottom of that page, under the caption "My Question(s)" you will find "questions or topics with unassigned points " Clicking that will give all, and only, your questions that still have unassigned postings.

Thanks on behalf of your Forum colleagues.

PS. - nothing personal in this. I try to post it to everyone with this kind of assignment ratio in this forum. If you have received a posting like this before - please do not take offence - none is intended!

Proost.

Have one on me.

jpe

[Wim Van den Wyngaert]

"The wizard is yet to see a single cases of INDEXF.SYS filling up on disks initialized on OpenVMS V6 or higher."

So, you might be the first. Did the disk come from pre-6 ?

Wim

[Jorge Cocomess]

No. Well, first time for everything, right?

Thanks,
J

[Hein van den Heuvel]

>> No. Well, first time for everything, right?

I doubt it.

I have yet to see a real error message, or useful data about the disk.

Just run a DFU report and stop speculating!

Hein.