logical set 60000, error -1

[calval35]

Hi,
I am getting this error " File System : Failed to resynchronize logical set 60000, error -1". I am assuming I have a bad disk?
But to replace a bad disk I need OS3.4 (At least that is my understanding)
If I am way off the mark here, please let me know. I don't have much experience with SNAP's. If I am not, I would really appreciate info on how to get 3.4.

thanks

Snap 4100
3.1.603
hardware:2.2.1
Bios:2.4.437

[blue68f100]

60000 is a raid5 partation.

You need to locate in the error log as to which drive is generating the error. Then replace that drive. Set as hot spare and it should then resync the set.

[calval35]

would I do this by entering debug mode? debug info?
You will have to excuse my SNAP ignorance.

Thanks again.

[blue68f100]

Once you get into the admin section there is gui at the bottom rt of the screen, above close.

or from debug try "in lo p -l" and "in de"

[calval35]

Great. Thanks.
I didn't see any disk or raid errors. File system error was reported though.
"File System Check : Failed to allocate 83841028 bytes for rcd_backlinks!!"

I will keep fumbling around.

Thanks for your assistance

[blue68f100]

Have you thought about resyncing the drives?

[calval35]

nope. how is that done?
Do you think upgrading the OS is good idea to?

Thanks again.

[blue68f100]

UP GRADEING IS NOT A GOOD IDEA WHEN YOU ARE HAVING PROBLEMS.

I think the cmd is "co de resync 6000"

[redrob]

What is the best course of action when this error occurs ? this is happening on a clients snapdrive and i am totally unfamiliar with these
TIA

BTW
It is a 4100 ,

[blue68f100]

That error is give when a drive is failing.

If you can still access the data, you should back it up and take the unit off line.

The course of action would be to look in the error log and see if you can determine which drive is failing. Replace the drive, set as hot spare, it shoud then rebuild the array.

[redrob]

where is the log that will be more specific about which it is only log i see is the "Server Log" , does the "-1" mean its the drive on channel 1

[blue68f100]

from debug "in de" will give details on devices
or "in lo p -1 " may help

60000 refers to the raid 5 array.

Drives are identified as 10000, 10007, 10010, 10017

[jontz]

To get to debug, in case you don't know, just add /debug to the end of the web address (ie, //(snapserver)/config/debug)

[calval35]

I can't figure out if there is actually anything wrong. The 4 drives are all listed. No errors?

Logical Device: 10006 Position: 0 JBOD Size (KB): 32296 Free (KB): 24424 Private Mounted
Label:Private Contains system files only
Unique Id: 0x5AB11EC664E1189C Mount: /priv Index: 0 Order: 0
Partition: 10006 Physical: 10007 FS Size (KB): 32768 Starting Blk: 515 Private
Physical: 10007 Drive Slot: 0 IDE Size (KB): 75067904 Fixed

Logical Device: 60000 Position: 1 RAID Size (KB): 220516096 Free (KB): 74287416 Public Mounted
Label:RAID5 Large data protection disk
Unique Id: 0x0D0BBF1266F233D8 Mount: /0 Index: 1 Order: 1
Partition: 10000 Physical: 10007 R 60000 Size (KB): 74524432 Starting Blk: 67806 Public
Physical: 10007 Drive Slot: 0 IDE Size (KB): 75067904 Fixed
Partition: 10008 Physical: 1000F R 60000 Size (KB): 74524432 Starting Blk: 67806 Public
Physical: 1000F Drive Slot: 1 IDE Size (KB): 75067904 Fixed
Partition: 10010 Physical: 10017 R 60000 Size (KB): 74524432 Starting Blk: 67806 Public
Physical: 10017 Drive Slot: 2 IDE Size (KB): 75067904 Fixed
Partition: 10018 Physical: 1001F R 60000 Size (KB): 74524432 Starting Blk: 67806 Public
Physical: 1001F Drive Slot: 3 IDE Size (KB): 75067904 Fixed

[jontz]

Looks similar to the page I get when I debug mine, doesn't seem to indicate any drive failure. If you go into the normal config screen, go to disk utilities, and then view disk status, you should see the raid 5 array status. If you click on the array icon it should show you any disk failures/problems. Let me know what you see when you do that. If there isn't a drive failure you might need to re-sync the drives.

[calval35]

here goes:

E File System : Failed to resynchronize logical set 60000, error -1 Disk 1 5/10/2006 1:17:43 PM
I File System : Extended Rights Backup for device 0x60000 has completed successfully Disk 1 5/10/2006 12:18:50 AM
I File System : Extended Rights Backup for device 0x60000 has begun Disk 1 5/10/2006 12:08:56 AM
I File System : Extended Rights Backup for device 0x60000 has completed successfully Disk 1 5/9/2006 12:18:25 AM
I File System : Extended Rights Backup for device 0x60000 has begun Disk 1 5/9/2006 12:08:31 AM
I File System : Extended Rights Backup for device 0x60000 has completed successfully Disk 1 5/8/2006 12:18:04 AM
I File System : Extended Rights Backup for device 0x60000 has begun Disk 1 5/8/2006 12:08:06 AM
I File System : Extended Rights Backup for device 0x60000 has completed successfully Disk 1 5/7/2006 12:17:26 AM
I File System : Extended Rights Backup for device 0x60000 has begun Disk 1 5/7/2006 12:07:41 AM
I File System : Extended Rights Backup for device 0x60000 has completed successfully Disk 1 5/6/2006 12:17:17 AM
I File System : Extended Rights Backup for device 0x60000 has begun Disk 1 5/6/2006 12:07:15 AM
I File System : Extended Rights Backup for device 0x60000 has completed successfully Disk 1 5/5/2006 12:16:42 AM
I File System : Extended Rights Backup for device 0x60000 has begun Disk 1 5/5/2006 12:06:51 AM
I File System : Extended Rights Backup for device 0x60000 has completed successfully Disk 1 5/4/2006 12:16:18 AM
I File System : Extended Rights Backup for device 0x60000 has begun Disk 1 5/4/2006 12:06:25 AM
E File System : Failed to resynchronize logical set 60000, error -1 Disk 1 5/3/2006 11:37:52 AM

Thanks a bunch as the plot thickens.

[blue68f100]

DISK 1 has a problem.

If you have another 80 gig, hd, replace disk 1, set as spare. And let it do its think. All snap os can handle this.

Do you have a copy of SpinRite by GRC. I would remove disk 1 and seen if spinrite can repair it.

If I have some free time, I may pull one of my 2200 drives out and use it as a test.

You do have the data backed up don't you ???????????????

[calval35]

great thanks. Yes data is backed up daily!!!
As far as spinnrite is concerned, no I don't.
In the mean time I will order another 80gb hd. I actually have 4 120's, how much of a hassle would it be tor rebuild it all together?

[blue68f100]

The 4100 has OS stored in flashram. Just install the new drives and turn it on. Give it 5 min to copy the OS to HD before you connect. Let it finish formating, then build you Raid5 array. What could be easier.

[jontz]

Raid 5 array will take about 4 hours to build, FYI. When it is done just copy all of your data back over to it.

[ShrewLWD]

Hi everyone,

I have a 4100 (4.0.860, HW 2.2.1, BIOS 2.4.437) that was in RAID 5 (4x120Gig Maxtor). One of the drives crashed, so I shipped it off to Adaptec. They sent me a 160gig WD. When it rebuilds, I get the 60000 error -1 message.
In the log its;
File System: Failed to resynchronize logical set 60000 Disk 60000

When I run in lo p -1 it shows it gets to 71% done on RAID5Resync then shows 3 identical lines;
DISK: req=0x515EC60 dev=0xc0003 fn=1 blk=0xA39AD70 sts=20
DISK: req=0x515EC60 dev=0xc0003 fn=1 blk=0xA39AD70 sts=20
DISK: req=0x515EC60 dev=0xc0003 fn=1 blk=0xA39AD70 sts=20

followed by;
read to cache failed (20)

in de shows all 4 drives with their respective assigned names; 10007, 1000F, 10017, 1001F but no errors or issues.

What am I missing here to determine a) if it's a drive issue and therefore b) which drive?

I shut it down and reboot it, which causes it to go through its 4+ hours of resyncing, but it still stops at 71%

Any assistance would be appreciated!


UPDATE: This thread (http://forums.procooling.com/vbb/showthread.php?t=13851&highlight=read+to+cache+failed) discusses it, and it appears dev=0xc0003 means drive 3? That's the replacement drive they sent me.

[blue68f100]

Can you currently read the data on the snap? If so back it up. If not what is the drive utility showing?

If the unit was upgraded it's possiable to have a different starting point. Snap changed the way they calculated the starting point. If you look at the "co de info" or "in de" closely the starting points should be the same. If not you have limited options. You need to pre-format the drive with the OS version used for the original ones.

Did the snap set the capacity to match the originals drives? If not use WD utility to set the size.

You may need to wipe the drives clean and start over.

[ShrewLWD]

We decommisioned this snap, and were in the process of reconfiguring it for a new job in our infrastructure when the drive crashed, so when I got the new drive I just broke the RAID and recreated it. We can do whatever we want to it at this point.
co de info shows the new drive does *not* have the same starting block.
So you are saying, break the raid, individually format the drive (using the format in the snap appliance?) then rebuild the raid?

When I built the raid with the new drive, it popped up saying like it was only going to use 119Gig of the drive to stay in compliance with Raid, so I assumed it formatted it at 119.

[blue68f100]

What you need to do is break the array. Then from debug issue "co de format XXXXX /reinit" . Where XXXXX is the drive ID's, 10000, 10008, 10010, 10018. Then recheck the starting points all should be the same and capacity may differ. If so the smallest drive must be in position 1.

[ShrewLWD]

OK, I broke the array, then sent that command to the 10010 drive (drive 3). While it's formatting, I did notice that upon breaking the array it formatted and reinitialized all the drives (or at least it said it was in the Disk Config view). Is the command you gave me going to do something different than what it did?

UPDATE: After running that command and re-running in de, the start block is still different than the others (the 3 originals are starting at 95927, this last one starts at 104776)

I don't see how formatting the others will make them jump up to that block, since he didn't bother to ask the others what block they are starting at.

Is there a different command to use?

[blue68f100]

The cmd I gave you is like a low level format. It ignors the previous settings, correcting anything that is not correct.

You need to to all drives. Either remove them all and use the mfg utility to format (quick zero of mbr will work). Then let the snap do a complete setup like all new drives.

The reason why all drives must be done.:
On some version of the os the starting point was determined by taking the number of sectors divided by 7. Then snap changed it to divide by 8.

[Phoenix32]

OK, I broke the array, then sent that command to the 10010 drive (drive 3). While it's formatting, I did notice that upon breaking the array it formatted and reinitialized all the drives (or at least it said it was in the Disk Config view). Is the command you gave me going to do something different than what it did?

UPDATE: After running that command and re-running in de, the start block is still different than the others (the 3 originals are starting at 95927, this last one starts at 104776)

I don't see how formatting the others will make them jump up to that block, since he didn't bother to ask the others what block they are starting at.

Is there a different command to use?



Use the command he gave you on ALL 4 DRIVES. This causes a full reinit of the drive (including start block), where normal format does not.

Early version of the SNAP OS and the later versions had different start blocks. This unit has probably had the OS updated since the original init.

[ShrewLWD]

I ran it on all, one at a time, rebooting so the command could run, and it still shows different start blocks.

[ShrewLWD]

Is there an option to have it run simultaneously on all of them?

If not, should I take them all out, attach them to a generic computer, and do a low level format on each of them?

[eschw95458]

When you type in "co de format 10000 /reinit" you should then get a message that says
"The requested operation will be executed after the next reboot.
Look in log for progress..."

DO NOT reboot type in the next one IE....."co de format 10008 /reinit" etc, etc,
Then when you have all the drives ready simply type r and hit enter

[ShrewLWD]

Cool! I'll try it right away.

I do have a question though..it appears as if 10000 and 10008 are designations of partitions, while 1000f and 10007 etc., are physical drives. Shouldn't I be be doing something to the physical drive, rather than a partition on a drive?

[ShrewLWD]

Unfortunately, it still starts 3 of them at 95927 and the new drive at 104776.

Anything else I can try?

[eschw95458]

Yes, you can pull all the drives and format them, it's been a while but I think you can do fat32, put them back in and power the unit back up. That will of course wipe everything.

[ShrewLWD]

will give that a try next! Thanks for you guys' patience! It's frustrating for me too, just wanting to get this thing back up and running.

[eschw95458]

If you look right under where it says Logical device it says Label. Then you can see where it says Drive1 2 etc.

[ShrewLWD]

Now one thing I read is that, because I'm going to be wiping drive 0, I will not have access to the admin screens for awhile, until it does it's internal formatting and setting up, correct?

[eschw95458]

It really shouldn't be too long, really not much longer than when you issued the /reinit cmd. It is possible to log in to the 4100 with no drives installed :-)

[ShrewLWD]

Hey hey..MN too? Cool!

So do a low level format, or just connect them to a PC and do a Windows format?

UPDATE: Man, I hope doing a quick format will suffice...doing a full format for each 120Gig HD is going to take forever! (I know, I have no patience! heh!)

[ShrewLWD]

*Sigh* OK, even after formatting all of them separately in a Windows machine, then putting them back, the in de shows the starting blocks off.

Low level format next?

[blue68f100]

The snap will take ~5-10 min before you will gain access to the admin.

Make sure the larger drive is not installed in position 1 or 2.

I have noticed on some units installing 1 drive at a time works.
You can also try adding the "/nocore" flag to the /init cmd.
Last but not least, use WD drive utility to read the size of your original HD then adjust the new drive to match.

[ShrewLWD]

Dang it! I did a low level format on drive 0, then put it back (only) and powered it up. It took awhile (as you said) to get to the admin window, but when I finally did, it showed it going to its original starting block (95927)
I can only hope that low level formatting the new drive will allow the snap to start at that same starting block.
I got Z & W, but they require true dos bootup, and I have been attaching these drives as USB to wipe them. I will need to monkey with my setup a bit to get them to be true local drives.

Update: hey blue, when you say dont put it in pos1 or 2, is that counting 0 1 2 3, or 1 2 3 4?

[ShrewLWD]

OK, this is getting a tad absurd! What is dictating the use of the starting block..the snap or the drive? I low-level formatted the new one, put it in slot 3 (drive bay 4, where it has not been before) and fired it up with newly ll-formatted drive 0 in its normal bay. Same thing...the new drive wants to start on 104776.

Should I send it back to adaptec and insist they send me a drive that starts on 95927?

It shouldn't be this hard to get something this trivial up and running! Why does it even care that they all need to start on the same block? Can it not keep individual tabs on each drive?
/rant off

[blue68f100]

Did you tried the following "co de format 10018 /reinit /nocore" ? Where the HD is in drive 4 position.

Also go to WD and download there drive utility. It will allow you to adjust the drive size. If you do not find it let me know and I will send it to your.

WHAT OS VERSION are YOU RUNNING ?????

[ShrewLWD]

My setup is
OS 4.0.860,
HW 2.2.1,
BIOS 2.4.437

I am running the /nocore option now.
I think I know what software you are referring to, but I don't understand how setting it to a size is also going to tell it to move the starting block. It seems to have remarkable resilience to any change.

UPDATE: OK, /nocore did move the starting block! Unfortunately, not to the same as the others! It's a smaller number, but still an even number (88472)

Should I now try reinit again without the /nocore ?

[ShrewLWD]

Just to make sure I'm not wasting you guys' time...I am not rebuilding any raid everytime. I simply follow your suggestions, then watch it format the drive and initialize it. I then check in de to see what the starting block is. I didn't think building a RAID was going to change what the suggestions you offered could not.

[ShrewLWD]

Nope, redoing the format without the /nocore took it back to 104776.

[blue68f100]

Use the mfg utility to change the drive size. Since all else has failed.

[snap-tech]

ShrewLWD

The following error means that the last drive has a an issue reading/writing to

DISK: req=0x515EC60 dev=0xc0003 fn=1 blk=0xA39AD70 sts=20
DISK: req=0x515EC60 dev=0xc0003 fn=1 blk=0xA39AD70 sts=20
DISK: req=0x515EC60 dev=0xc0003 fn=1 blk=0xA39AD70 sts=20

This line tells you what drive. The last digit = the drive. 0-3.

dev=0xc0003


Your options are to either send drive back if this is the drive you just received. Or to run some kind of drive repair util to locate bad sectors and mark them bad, then retry to use drive in snap.


I will post another message in an hour or so, explaining when you need to worry about starting blocks and when not to worry.

Douglas Snap-Tech

[snap-tech]

Do you know if the snap was running a 2x version or a 3x version when the raid was originally built ( in other words the first time).

[ShrewLWD]

Well, geat news late last night! I had wiped all the drives, and was resetting them up in the original RAID5 configuration so that when it crashed I would have up to date logs for an adaptec support call I was going to make this morning. It got to the 71% mark... then kept going! It actually completed the process and gave a greenlight to the setup!

Of the things different, all drives were low level formatted, and the newer, larger drive was moved from bay 3 (drive 2) to bay 4 (drive 3)
The starting block is still off, but it's working.

A really big thank you to you guys for assisting, even on weekends!!!

snap-tech, yes, knowing when starting blocks are an issue would be very useful info! Please do share!

As far as updates go, we only went from the version prior to .860 (.805?), to the 860 update, due to the Windows SMB patch issue.

[snap-tech]

How did you wipe all drives? Did you put them into pc and write 0's to all sectors?

The drive that is in bay 4 which is actually drive 3 because we start counting from 0, does have a bad or going bad sector.

I would check the debug log using the following cmd at least once a week and look for the following errors

command = i l p -1

error = DISK: req=0x515EC60 dev=0xc0003 fn=1 blk=0xA39AD70 sts=20

dev=0xc0003 = the last digit = drive# (0 - 3)

blk=0xA39AD70 = the block number.

It will take me a few minutes to explain the starting blocks, and will have some time available later today. sorry for delay on that.

Douglas Snap-Tech

[ShrewLWD]

Snap-tech I used HDDGuru's Low level Format Tool.
I will certainly take a look at the logs very often then!

Some questions though... if it didn't catch it during the setup, what is going to catch it now? Does it have a built-in drive checker? If so, can I force it to check now?

[blue68f100]

Snap-tech explained to me that the starting point is only in play if the array was built using OS v2 and was later upgraded to v3 up. So the starting block being at different location on the larger drive is OK as long as it is not in Position 1 (10000).

ShrewLWD did your low level utility do a surface check. If not use a utility like SpinRite to check and block out all bad sectors. You do not need seek and sector errors on raid5 arrays.

[dtebbe]

I hate to hijack the thread, but I'm having the same problems. I installed 4 new 80g drives and when I try to build the RAID it gets to about 30% of the spare and then fails. Here is the log :

11/11/2008 19:21:55 77 D SYS | RAID5Resync on array 0: 29% done
11/11/2008 19:23:45 77 D SYS | RAID5Resync on array 0: 30% done
11/11/2008 19:25:31 77 D SYS | RAID5Resync on array 0: 31% done
11/11/2008 19:25:51 77 D NET | INET: sendit: send err 39
11/11/2008 19:27:16 77 D SYS | RAID5Resync on array 0: 32% done
11/11/2008 19:28:40 77 D SYS | DISK: req=0x2158550 dev=0xC0003 fn=1 blk=0x31608F0 sts=20
11/11/2008 19:28:40 77 D SYS | IDE: A 48 bit LBA capable Hard Drive was detected ...
11/11/2008 19:28:40 77 D SYS | IDE: 48 bit LBA is not supported on 4100 and 12000 platforms.
11/11/2008 19:28:40 77 D SYS | IDE: Setting the hard drive size to maximum addressable by 28 bits: 76319 MBytes.
11/11/2008 19:28:44 77 D SYS | DISK: req=0x2158550 dev=0xC0003 fn=1 blk=0x31608F0 sts=20
11/11/2008 19:28:44 77 D SYS | IDE: A 48 bit LBA capable Hard Drive was detected ...
11/11/2008 19:28:44 77 D SYS | IDE: 48 bit LBA is not supported on 4100 and 12000 platforms.
11/11/2008 19:28:44 77 D SYS | IDE: Setting the hard drive size to maximum addressable by 28 bits: 76319 MBytes.
11/11/2008 19:28:44 77 D SYS | DISK: req=0x2158550 dev=0x80003 fn=1 blk=0x31608F0 sts=20
11/11/2008 19:28:44 77 D SYS | RAID5Resync on array 0: read to cache failed (20)
11/11/2008 19:28:44 77 D SYS | RAID5Resync on array 0: failed
11/11/2008 19:28:44 77 E L01 | File System : Failed to resynchronize logical set 60000, error -1
11/11/2008 19:38:06 77 D NET | Network transmission errors : More than 1 percent re-transmissions reported by TCP/IP.


Model -4000 (Dell 705N w/Snap software upgrade)
Software - 3.4.805 (US)
Hardware - 2.2.1
Server # - 508021
BIOS - 2.4.437

Any ideas?

TIA

DT

[blue68f100]

Bad power supply would be my guess ......

Use a PC power supply to power up the HD's and see if it works to confirm.