Jump to content

Error -1019 (not enough resources)--What kind? How much?


awnews

Recommended Posts

Any particular *kind* of "resources" and how much of XXX is needed since it's **clearly** not an issue of disk space? The destination drive is a 120GB NTFS-formatted drive with 63GB free. This backup set is only consuming 16GB of that and this incremental backup only added 2.6GB. And I have trouble believing that it's memory since other backups backuped up the same directory (about 30GB/109K folders/297K files) a day before and a day after. Machine is XP SP1, 1G RAM, several large NTFS drives. Destination is an IDE drive on an XP SP1 1G RAM machine across a 100Mbps LAN.

 

 

+ Normal backup using V_Data_Backup at 6/13/2004 9:02 PM

To Backup Set V_Data_Docs...

 

- 6/13/2004 9:02:32 PM: Copying Docs on Drive V (V:)

Can't save Snapshot in V_Data_Docs, error -1019 (not enough resources)

6/13/2004 9:39:36 PM: Comparing Docs on Drive V (V:)

6/13/2004 10:13:19 PM: 1 execution errors

Completed: 1969 files, 2.6 GB, with 46% compression

Performance: 109.0 MB/minute (101.9 copy, 117.2 compare)

Duration: 00:59:14 (00:11:34 idle/loading/preparing)

Link to comment
Share on other sites

Hi

 

Generally this kind of error means memory but it could be caused by disk space as well. For snapshot creation Retrospect requires large blocks of continuous memory. It may be that it couldn't that on the day the errors occured. Where is the catalog files stored? How about the Retrospect preference files? How much space is free on those drives? Have you tried defraging the machines involved?

 

Thanks

Nate

Link to comment
Share on other sites

 

> Generally this kind of error means memory but it could be caused by disk space as well. For snapshot creation Retrospect requires large blocks of continuous memory. It may be that it couldn't that on the day the errors occurred.

 

"continuous memory" == disk space I assume. Are you saying Retro wants *continuous* memory somewhere (dest. drive? local C: drive) for saving snapshots *and* can't have this (disk) memory fragmented?

 

 

> Where is the catalog files stored?

 

File Backup to dest. harddrive, so in the File.

 

 

> How about the Retrospect preference files?

 

On the main C: drive (I assume). Backing up from another IDE drive (V:) on the same PC as C to a dest (W:) drive on a LAN PC.

 

 

> How much space is free on those drives?

 

63GB free out of 120GB harddive (as noted) on dest (W:) drive. Source V: drive has 30GB free out of 80GB. Local C: drive has 44GB out of 60GB free.

 

 

> Have you tried defraging the machines involved?

 

The destination drive does occasionally get degragged (scheduled PerfectDisk) but hasn't for a couple of months. The backup IDE drive (W:) is very unfraggmented (no programs or OS, just backups) *except* for the three Retrospect File backups that are saved to this disk, which do have some degree of (normal) fragmentation.

 

 

FYI, the backup ran again over the weekend under the same conditions without reporting any errors. No defrag has been run since the last time when the backup logged the "error -1019 (not enough resources)."

Link to comment
Share on other sites

Hi

 

My understanding is that retrospect needs continuous blocks of memory. I beleve this applies to RAM as well as virtual memory. The resources error could just indicate that Retrospect was unable to get the kind of memory it needed on that particular backup.

 

Thanks

Nate

Link to comment
Share on other sites

  • 3 weeks later...

I'm seeing the "not enough resources" message happen more frequently. It seems to be assocated with backing *from* a particular subvolume (with a lot of file), not *to* one of several destination volumes. As noted in my original post, the included "Can't save Snapshot" suggests that the data may be getting saved but that the snapshot isn't.

 

Since I have a gig of RAM (more that most PC users) on this XP SP1 PC, it's not reasonable to just say "add more RAM"--Dantz needs to find a way to deal with this situation (e.g. breaking up the snapshot into smaller chunks). And I don't have any "resources" on my PC--I have RAM, disk, cache, etc.--so you need a better message here clearly stating what Retro is running out of, how much it needs, etc.

Link to comment
Share on other sites

Hi

 

I can assure you, memory management is high on our priority list. The bummer is that it will require some serious work which boils down to more time.

 

How many files are on this disk? Is the error more likely to occur when you do a recycle backup or when you do normals?

 

Do you have the backup NTFS security option enabled? That will make the snapshot larger.

 

How large is the swap file set to be? Try increasing it if you can.

 

How large is the snapshot for this volume when the backup is successful?

 

Thanks

Nate

Link to comment
Share on other sites

 

> How many files are on this disk? Is the error more likely to occur when you do a recycle backup or when you do normals?

 

Just under 300K files /110K folders (28GB total) in the subvolume being backed up from a second drive in the system. I'm getting the error (every time now) on all backups including normals from this subvolume. And it seems to be caused by the snapshot (so same size regardless of normal/recycle).

 

> Do you have the backup NTFS security option enabled? That will make the snapshot larger.

 

This is set up with the default, i.e. [x] Back up server security informaiton and [ ] (unchecked) Back up workstation secuity information. This is on an XP Pro SP1 PC.

 

 

> How large is the swap file set to be? Try increasing it if you can.

 

With 1G of RAM, the swap is set from 1.5G (min) to 3G max.

 

> How large is the snapshot for this volume when the backup is successful?

 

I don't know. Where would I find the snapshot size?

 

Thanks

Nate

Link to comment
Share on other sites

Hi GoAWest

 

 

 

The shapshot size is listed in the log after a snapshot is saved.

 

 

 

Something is definitely not right here. 300K files should be no problem at all.

 

Can you try a backup to a new set on the local disk using the "no files" selector. More than anything I am interested to see if it works at all. Open task manager and keep an eye on memory useage when this runs. That will tell us how much memory is being used during snapshot creation.

 

 

 

Is this a "file" backup set or a "disk" backup set? If it is a disk set, is the catalog local or on the remote machine?

 

 

 

The error is "can't save snapshot" which leads me to believe the snapshot has been built it just isn't being sent over the network properly.

 

 

 

Thanks

 

Nate

Link to comment
Share on other sites

> The shapshot size is listed in the log after a snapshot is saved.

 

Thanks. I was trying to look somewhere in the backup file itself. The snapshot for this backup source is about 330MB. This is true for versions of the failing script (as of a couple of weeks ago) and for backups of the same source to another another drive on this same system. The failure is when backing up & snapshoting from this "V" drive to a location over the network.

 

 

> Is this a "file" backup set or a "disk" backup set? If it is a disk set, is the catalog local or on the remote machine?

 

It's a file backup. I've given up on disk backup sets due to the inabilty to reasonably specify a destination (e.g. in a drive and a folder path).

 

 

> The error is "can't save snapshot" which leads me to believe the snapshot has been built it just isn't being sent over the network properly.

 

That does jive with the fact that another backup script on this same PC seems to work from the same source but to a *local* destiation (another harddrive). It's only the backups from this source to destinations over the network that are showing failing snapshots.

Link to comment
Share on other sites

Hi

 

How much memory and what is the OS on the client machine? If you create another file backup set and backup this volume do you get the same problems? Do you have the "backup security information from servers/workstations" option enabled? That will add to the size of the snapshot.

 

When Retrospect saves the snapshot it has to roll it into a temp file that contains updated catalog information. This file is stored in the Retrospect preferences folder on the C drive.

 

One theory (albeit a long shot) is that a driver or I/O problem is causing a failure when Retrospect tries to write the snapshot to the catalog. At that point Retrospect sees the disk as full and fails. This could also pertain to the C drive where the temporary file containing the snapshot is stored.

 

Have you tried rebuilding the catalog for this file set?

 

Thanks

Nate

Link to comment
Share on other sites

 

> How much memory and what is the OS on the client machine?

 

Source machine: WinXP Pro SP1, 1G RAM

C: 60G drive, 40G free, NTFS

V: 80G drive, 30G free, NTFS

Dest. machines: WinXP Pro SP1, 1G RAM across 100M LAN to 120G drive, 50G free, NTFS

Linux box (disk & RAM?) across 100M LAN

 

> If you create another file backup set and backup this volume do you get the same problems?

 

Haven't tried it with *another* backup set, but it's failing with two dest. as noted. So the common issue is the source.

 

 

> Do you have the "backup security information from servers/workstations" option enabled? That will add to the size of the snapshot.

 

Servers option Yes, Workstation option No.

 

 

> Have you tried rebuilding the catalog for this file set?

No, but it's two different backup sets with two different destinations.

Link to comment
Share on other sites

No problem. I've been reading the forum long enough not to get annoyed at the lags.

 

The general "Software data compression" option is turned on. There's no specific compression option (??) for the catalog (in this File backup).

 

Overall, this is an odd mix. I have several backups from the same "V:" drive, some for the entire root docs folder (not /Docs & Settings, just a V:/Doc/s folder at the root of the C: drive) and others that just back up subfolders in that directory (e.g. V:/Docs/one/ or /Docs/two/, etc.). Some back up from the V:/Docs/x sources to another drive on the same PC (no errors, no "out of resources"). One backs up up from V:/Docs/x to an Ximeta NetDisk (over Ethernet with odd non-TCP protocol), no errors or "oor" messages. The two that give me the "oor" errors (only on the snapshot save) are both backups *to* devices over the network (TCP/IP), one an XP PC with a shared drive and one a Samba drive. However both of these destination have plenty of memory and disk space.

Link to comment
Share on other sites

re:

 

>The general "Software data compression" option is turned on. There's no specific >compression option (??) for the catalog (in this File backup)

 

There should be a catalog compression option when you look at properties of the backup set, and then look at the 'options' tab, which is far away from the script options to compress.

 

Good luck

Link to comment
Share on other sites

Thanks, I found it.

 

1) No, the catalog is not set up for compression.

 

2) IMHO, it's pretty dumb to have almost all the options for the backup set in the script Options tab but then have this oddball Catalog compression option under some info for the backup set. Put the Catalog compression option setting under the script Options->Catalog tab where it belongs...

Link to comment
Share on other sites

Here's the Retro log for one backup failing with the "error -1019 (not enough resources)" message. I turned the "trees and volumes" debug logging up to 6 and the Network logging up to 4.

 

I had rebooted the machine yesterday. I've left OFB on. The first "error -1017 (insufficient permissions)" is our old friend that I believe is the same issue as reported previously--"something" is turning off or not correctly enabling (set to Auto) the Volume Shadow Copy service and the RP message ("insufficient permissions") is bogus. I've manually Started this service for next time.

 

The "error -1019 (not enough resources)" then follows, with whatever surrounding messages are captured in the log.

 

 

 

+ Normal backup using V_Data_to_LS1_Server at 7/28/2004 5:03 AM

To Backup Set V_Data_Backup...

VssWEnumWriterMetadata: Can't exclude writers on Windows XP, API call not supported

VssWSnapVolume: DoSnapshotSet failed, winerr -2147212538, error -1017

VssWCleanupMetaSet: DeleteSnapshots failed, winerr -2147212536, error -1017

Can't use Open File Backup option for Projects on Drive V (V:), error -1017 (insufficient permissions)

 

- 7/28/2004 5:03:49 AM: Copying Projects on Drive V (V:)

lvtrRescan: avail - minstart = 429,078,936, minstart = 163,840, minmem = 98,304

VssWEnumWriterMetadata: Can't exclude writers on Windows XP, API call not supported

Excluded 0 files, zero K

lvtrTaskRescan: scanning: 227.547 of 355.0 secs (64.09%) for 1 ops (227,547,000.0 uSec/op)

Can't save Snapshot in V_Data_Backup, error -1019 (not enough resources)

OutlookFileListappl.gifReinit: found 0 files that Outlook may be locking.

Could not get Outlook version, so turning off Outlook agent.

0 outlook files. 0 outlook procs (0 were stopped, 0 restarted).

FEncStart: done

FEncEnd: begin

FEncEnd: starting to wait

FEncEnd: still waiting

FEncEnd: done

FEncStart: done

FEncEnd: begin

FEncEnd: starting to wait

FEncEnd: still waiting

FEncEnd: done

7/28/2004 5:17:47 AM: 1 execution errors

Completed: 662 files, 157.0 MB, with 81% compression

Performance: 67.7 MB/minute

Duration: 00:12:06 (00:11:25 idle/loading/preparing)

 

FEncEnd: not started

FEncEnd: not started

FEncEnd: not started

FEncEnd: not started

vlocQuitHook: DK_DRMODE is false.

FEncEnd: not started

FEncEnd: not started

FEncEnd: not started

FEncEnd: not started

Quit at 7/28/2004 5:18 AM

Link to comment
Share on other sites

Thanks for the info.

 

I'll see what I can find out about this.

 

It is really odd that the Volume shadow copy service will fail like that. This is a regular windows service so it should work just fine. I wonder if there is a problem with services in general on XP - that might explain some of the Retrospect launcher service problems as well...

 

Nate

Link to comment
Share on other sites

Sorry ... I just have to chime in here with a whine about the incredible poor error handling in Retrospect. Error codes (with negative numbers) are from 1960 ... why can't retrospect give clear, English errors that describe what's going on. If the program is running out of memory, disk space, or gets a driver errors say so. Don't give messages of the sort "an error occured, sorry".

 

Also, what is "contigous memory". All PC since about 1995 have used virtual memory to map physical memory into what appears to be contigous memory. I think Retrospect is showing its Macintosh roots by a) blaming the OS and B) using unsupported system work-arounds.

 

I've been a fan of Retrospect (and have used it on the Mac) since the 1990... but the program (and pricing) is due for a serious rework.

Link to comment
Share on other sites

I'll get back to the issue that started this thread ("insufficient resources") but right now I'm fighting with "Can't use Open File Backup option for Projects on Drive V (V:), error -1017 (insufficient permissions)." After manually staring that service yesterday, I see three scripts that had reported this error last night. I know that Retro starts and stop the service, so it's not unusual to see it with a "stopped" status after Retrospect has run. This is seen in the system Event Properties log ("VSC successfully sent a start control" and "VSC service entered the running state.").

 

*However* I'm also seeing the following errors in the Applications Event Log (correlated with the times of my scripts):

 

Event Type: Error

Event Source: VSS

Event Category: None

Event ID: 12289

Date: 7/29/2004

Time: 1:15:58 AM

User: N/A

Computer: HWSERVER

Description:

Volume Shadow Copy Service error: Unexpected error DeviceIoControl(000001D8,0x0053c008,00269900,0,002688F8,4096,[0]). hr = 0x80070020.

 

For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.

Data:

0000: 49 4e 43 49 43 48 4c 48 INCICHLH

0008: 33 36 32 00 00 00 00 00 362.....

0010: 53 50 52 51 53 4e 50 43 SPRQSNPC

0018: 31 36 30 00 00 00 00 00 160.....

 

---------------------------------------------

Event Type: Error

Event Source: VSS

Event Category: None

Event ID: 12289

Date: 7/29/2004

Time: 2:37:04 AM

User: N/A

Computer: HWSERVER

Description:

Volume Shadow Copy Service error: Unexpected error DeviceIoControl(000001C0,0x0053c008,00269990,0,00268988,4096,[0]). hr = 0x80070057.

 

For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.

Data:

0000: 49 4e 43 49 43 48 4c 48 INCICHLH

0008: 33 36 32 00 00 00 00 00 362.....

0010: 53 50 52 51 53 4e 50 43 SPRQSNPC

0018: 31 36 30 00 00 00 00 00 160.....

 

--------------------------------------------

Event Type: Error

Event Source: VSS

Event Category: None

Event ID: 12289

Date: 7/29/2004

Time: 3:12:42 AM

User: N/A

Computer: HWSERVER

Description:

Volume Shadow Copy Service error: Unexpected error DeviceIoControl(000001C0,0x0053c008,00269990,0,00268988,4096,[0]). hr = 0x80070057.

 

For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.

Data:

0000: 49 4e 43 49 43 48 4c 48 INCICHLH

0008: 33 36 32 00 00 00 00 00 362.....

0010: 53 50 52 51 53 4e 50 43 SPRQSNPC

0018: 31 36 30 00 00 00 00 00 160.....

 

 

 

At the moment I've turned on an Automate script to Start the VSC service (if it's not already running) every few hours and will see what happens. However, this isn't looking like a start/stop issue but a problem with the VSC itself. And it doesn't look *exactly* like the issue (different error message) reported in the other "error 1017" thread.

 

http://forums.dantz.com/ubbthreads/showflat.php?Cat=&Board=professional&Number=42806&page=0&view=collapsed&sb=5&o=&fpart=all

Link to comment
Share on other sites

Hi

 

A wild idea but I wonder if the insufficient resources error is caused by a failure of the VSC service during the backup? That could explain the vague nature of the error message at least.

 

Do these errors occur in the event log when you manually start the service?

 

Thanks

Nate

Link to comment
Share on other sites

Hi,

 

I have this very same problem. It has been reported to Dantz, but no resolution so far. My -1019 come and go. The machine has 24gb of ram running w2k+3 entrpirse server. Their are 10x 140gb drives. We are using a NEO 4000 SDLT running 4 scripts at the same time. It works great with average transfer rates of over 1gb/min.

 

The storageset are about 6mb in size compressed. We do new media backups each night. The pagefile is 4092gb, the largest it can be without having multiple page files. We don't back up many files, but the files are large and some are zipped. One volume may have 105gb of data, but only 123 files.

 

I am backing up server security, not workstation security. It is Retrospect 6.5 latest build (350) and 5.1 rdu. A restart will normally fix the issue. When you check the ram before the restart only 580mb seems to be used.

 

I had two days of errors this week and the next night they went away. The scripts are all the same, just duplicated to run on different days.

 

We were a bit shocked to see this error when we had 24gb of ram. Did anyone ever get a real answer? All the harddisk have over 10gb spare.

 

Best wishes

 

Michael Curtis

Link to comment
Share on other sites

Hi,

 

Does this error occur when you have the "backup open files" option disabled? Do you have any errors in your event logs in regard to the Volume Shadow Copy service?

 

backing up a few large files should require only a minimum amount of memory and disk space. I suspect there is something else going on here.

 

Thanks

Nate

Link to comment
Share on other sites

  • 2 weeks later...

definently not a disk space issue... I am backing to an external USB 250G drive with nothing on it except from the back up that wont do the snap shot... so it might have to do with the amount of data on my c drive as the D drive save the snapshot in the same backup.

 

I have been playing around and moved 52 meg of data 400k files from the drive and now it seems to work OK... It apears there is some magic unknown limit.

 

All my local drives have at least 100 gig free...

 

What if anything can I use to check to see if it is some weird IO problem

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...