Pro/Forums

Pro/Forums (http://forums.procooling.com/vbb/index.php)
-   Snap Server / NAS / Storage Technical Goodies (http://forums.procooling.com/vbb/forumdisplay.php?f=82)
-   -   Guardian OS units (http://forums.procooling.com/vbb/showthread.php?t=14105)

blue68f100 05-08-2008 05:00 AM

Re: Guardian OS units
 
I dought that the HD will contain the OS, Andy generally full test the drives before selling any thing. So the drives will be clean but in perfect working order.

Phoenix32 05-08-2008 12:25 PM

Re: Guardian OS units
 
Yup, I use MHDD from the UBCD to check every sector on a drive for weak or bad sectors before I even think about putting a drive in a RAID 5. Or befor a drive ever leaves me to someone else. I also use Spinrite 6 for serious hardcore testing as required. I have a good reputation and want to keep it that way. :nod:

z0ner 05-28-2008 07:42 PM

Re: Guardian OS units
 
I've got a barely used 640GB 4200 that I'll be selling very soon. It's been in storage for almost a year and I believe has one of the newer (if not the newest) GuardianOS on it. I'll be powering it up to test shortly.

EDIT: Updated to GuardianOS 5.0.125. I might turn her into 2TB (4x500GB) , I haven't decided yet.

242 07-04-2008 08:29 AM

Re: Guardian OS units
 
I just got two 18000 with 8x250 each running os 3.2.019 anyone know what version of linux this thing is based off of?

-242

blue68f100 07-04-2008 09:08 AM

Re: Guardian OS units
 
I know v4 is RedHat Entriprise. At one time they were using Gentoo very early on. So my guess on v3 is RedHat Enterprise. V3 ( up to v 4.1) allows you to use Putty to SSH in to the shell. V4.4 has moved a lot of functions to the debug page. and requires a shell cmd to gain access.

242 07-04-2008 12:17 PM

Re: Guardian OS units
 
Quote:

Originally Posted by blue68f100
I know v4 is RedHat Entriprise. At one time they were using Gentoo very early on. So my guess on v3 is RedHat Enterprise. V3 ( up to v 4.1) allows you to use Putty to SSH in to the shell. V4.4 has moved a lot of functions to the debug page. and requires a shell cmd to gain access.

Interesting, I figured it would be common knowledge. I decided to take a look under the hood and I am find a few things that would lead me to believe it has a redhat base but I guess the labeled it Quantum Linux

Here is the contents of the release file in /etc
sh-2.04# more quantum-release
Quantum Linux IDE NAS (Opus)

Phoenix32 07-06-2008 10:54 AM

Re: Guardian OS units
 
The newer 5.0.125 looks like it is SUSE.

yfpang 07-15-2008 01:37 AM

Re: Guardian OS units
 
Dear All,

I'm not sure if I'm posting my question on the correct forum here, as currently I'm having a Quantum SnapServer running on Guardian v2 which is having a problem to boot up.
As from the serial output it seem like it's running fine in the beganing but later it comes to kernel panic.

As for testing I've tried to see if the OS reside in the HD or Flash and it's being reside in the HD.

Hope that there is some people here that may have seen this problem and resolve it before.


Thanks
Pang

blue68f100 07-15-2008 03:53 AM

Re: Guardian OS units
 
It would be nice to know which model you have, some of the early ones could run either OS depend bios/firmware size. V2 make it a very old unit. My first guess would be that it may need cleaning, unit over heating, maybe a cooling fan failure. Then I would suspect a HD problem.

yfpang 07-16-2008 01:01 AM

Re: Guardian OS units
 
1 Attachment(s)
Hi Blue68f100,

The model that currently i have is guardian 4400, which is running on Guardian OS version 2.1.003. As for the unit it's running fine until one of the administrator change the host name and restart the unit.

For this version do you thing it has something to do with the OS image? As the following is part of the serial output from the server :

Hope to hear from you soon

Thanks
Pang

blue68f100 07-16-2008 03:57 AM

Re: Guardian OS units
 
If you have a copy of SpinRite, I would shut it down and run it on all the drives. If there is a drive problem it will correct it. Otherwise you may need to reload the GSU file. It's has an error on loading one part. Like maybe all the markers did not get changed.

Have you checked to see if the bios settings have not changed? May be time to replace the Battery. But this is down Phoenix32 line.

Quote:

EXT2-fs warning: checktime reached, running e2fsck is recommended

VFS: Mounted root (ext2 filesystem).

kmod: failed to exec /sbin/modprobe -s -k char-major-10-152, errno = 2

kmod: failed to exec /sbin/modprobe -s -k char-major-10-152, errno = 2

/dev/Qhwmon: No such device
/dekmod: failed to exec /sbin/modprobe -s -k char-major-10-152, errno = 2

v/Qhwmon: No suckmod: failed to exec /sbin/modprobe -s -k char-major-10-152, errno = 2

h device
/dev/Qhwmon: No such device
/dev/Qhwmon: No such device
Using /lib/modules/i2c-core.o
i2c-core.o: i2c core module

Using /lib/modules/i2c-dev.o
i2c-dev.o: i2c /dev entries driver module

Using /lib/modules/i2c-proc.o
i2c-proc.o version 2.6.1 (20010825)

Using /lib/modules/i2c-piix4.o
Install SMBus adapter driver (i2c-piix4)

i2c-dev.o: Registered 'SMBus OSB4/CSB5 adapter at 0580' as minor 0

i2c-piix4.o: PIIX4 bus detected and initialized

Using /lib/modules/Qhwmon-Modulekmod: failed to exec /sbin/modprobe -s -k char-major-10-152, errno = 2

.o
kmod: failed to exec /sbin/modprobe -s -k char-major-10-152, errno = 2

/dev/Qhwmon: No such device
/dekmod: failed to exec /sbin/modprobe -s -k char-major-10-152, errno = 2

v/Qhwmon: No such device
/dev/Qhwmon: No such device
Qhwmon: proc entries installed OK

Qhwmon: Hardware monitor driver successfully loaded as Manta driver

Using /lib/modules/mtdcore.o
Using /lib/modules/mtdchar.o
Using /lib/modules/chipreg.o
Using /lib/modules/gen_probe.o
Using /lib/modules/cfi_probe.o
Using /lib/modules/cfi_cmdset_0001.o
Using /lib/modules/amd_flash.o
Using /lib/modules/quantumsnap.oGuardian: Found no Flash device


Using /lib/modules/mii.o
Platform = MANTARAY
Using /lib/modules/qinfo_i2c.o
Using /lib/modules/qinfo.o
qinfo: version 2.2 (20020709)

mdctl: fail to stop array /dev/md100root: No such device
starting raid: md100root containing drive /dev/hda2
mdctl: /dev/md100root has been started with 3 drives.
mdctl: fail to stop array /dev/md101swap: No such device
starting raid: md101swap containing drive /dev/hda5
mdctl: /dev/md101swap has been started with 4 drives.
>> root FS exists
***** Running on HARD DISK *****
Not on Clipper: Returning from Linuxrc
Kernel panic: No init found. Try passing init= option to kernel.

Oops: 0002

CPU: 0

EIP: 0010:[<c0230009>] Not tainted

EFLAGS: 00010206

eax: 00000007 ebx: c02f7ee0 ecx: 00000000 edx: 00000488

esi: 00000000 edi: 00000000 ebp: 0008e000 esp: dfde7f84

ds: 0018 es: 0018 ss: 0018

Process swapper (pid: 1, stackpage=dfde7000)

Stack: c02f7ee0 00000000 00000000 0008e000 00000246 00000000 c02304de 00000000

00000000 c02f7ee0 c011c9ee c02f7ee0 00000000 00000000 c02a8900 0000000e

c0105000 c0113c6a c0317f14 00000000 00000000 00000000 c02a88f6 c010539a

Call Trace: [<c02304de>] [<c011c9ee>] [<c0105000>] [<c0113c6a>] [<c010539a>]

[<c01056f3>]

Code: ff 01 0f 8e 25 05 00 00 31 c0 5b 5e 5f 5d 59 5a c3 89 f6 31

<0>Kernel panic: Attempted to kill init!


Phoenix32 07-16-2008 06:34 AM

Re: Guardian OS units
 
After reading all that GREEK, which gave me a headache, :hammer:

It looked to me like the OS is lost, confused, FUBAR, or just having a real bad day, which in turn makes me think Hard Disk, somewhere in the initial 10GB of a drive.

David, it could be a BIOS setting, but I think it's improbable (low odds). When that happens with these, they like to not boot at all until you go in and fix 'em.

I would look to see if you have a failing/failed drive somewhere in that mess and hope your OS is not hosed.

-AND-

David, wasn't the old v2 OS prone to confusing itself by not deleting things it should? I can't remember, but that is your department (punt).

blue68f100 07-16-2008 11:43 AM

Re: Guardian OS units
 
I think when they changed the name it did not complete the change to the links when it was rebooted. Inturn scrambling it's brain, the new pointer is not set completely. If the drives test out with SpinRite , he probably should do a clean install.

yfpang 07-16-2008 09:31 PM

Re: Guardian OS units
 
Hi blue68f100 and Phoenix32,

Thanks for the advise but I'm still very confuse as to what should i do from now.
Should i get 1 new HD and place it in there just to try booting it up or use the gsu file that I have to make the upgrade.

Furthermore do you think it's possible to copy the OS from the drive since it reside in it's own area that is not being effected by the RAID.

Hope to hear from you and Sorry for the greek word :laugh2:

Thanks
Pang

blue68f100 07-17-2008 03:37 AM

Re: Guardian OS units
 
The OS resides on all drives (raid1), part of the watch dog feature.

If you have a copy of spinrite, shut the unit down and run it on all of the drives. Then put them back in in the same order they were removed.

If you have the GSU ( which is the OS Image File) you can try the diaster recovery and reload the OS. This will not impact the data unless you select clean install.

Last if nothing else works you will need to wipe it clean with a clean install.

so
1st spinrite on all of the drives.
2nd reload the OS
3rd and last resort is a clean install ALL DATA WILL ME LOST.

yfpang 07-18-2008 01:51 AM

Re: Guardian OS units
 
Hi blue68f100,

But from the previous forum info, i though we can't use the GSU file to fix or reload the OS?
As in order to use the GSU file we may need a working OS first or the system must be able to boot until we are able to have access to the unit.

And for fresh Installation how do we go about it? Should we be looking for another unit of Guardian with working OS and load the new HD into the unit for it to sync the OS over to the HD? Or is there a way for us to load the OS into the fresh HD directly?

Thanks
Pang

blue68f100 07-18-2008 04:50 AM

Re: Guardian OS units
 
I would run SpinRite on the drives. More than likely there is marker wrong an it will repair it.

You should be able to get in to the recovery console. This is an option on v3 on up, have not looked at v2. Watch the terminal screen, there should be an option during the boot process. If you can gain access to the shell, use the "touch /nopiviot" <enter> then "reboot" <enter> This will force the unit into recovery console and will wipe out all data. "touch" with out the /nopiviot may just gain access to the recovery console.

With the watchdog bios you need to give the unit an hour on startup. The os is suppose to report back to the bios. If it does not after 5 min it should reboot using the next HD continuing through all of the drives till it finds one working.

If you decide you can do with out the data, You need to test all of the drives for a working OS. More than likely it's only main boot drive that is bad.

Failing a HD or 2 should also envolk the diaster recovery console. From there you can reload or do a clean install.

slkj 07-29-2008 11:52 AM

Re: Guardian OS units
 
Hi,

Does anyone know of a source for a replacement power supply and fans for a SNAP 4200? We have a couple of 4200s that are working well but want to have backups of the things that might fail. Adaptec says they don't have any.

Thanks,
Steve

bitor 07-29-2008 01:06 PM

Re: Guardian OS units
 
You are going to have to find out the serial/part number for the fans inorder to find a replacement for them. It's on a sticker on the fan. Then do a google to find it.
Most of the power supplies are from http://www.zippy.com/ but yours may be diff.

bitor


Quote:

Originally Posted by slkj (Post 170193)
Hi,

Does anyone know of a source for a replacement power supply and fans for a SNAP 4200? We have a couple of 4200s that are working well but want to have backups of the things that might fail. Adaptec says they don't have any.

Thanks,
Steve


blue68f100 07-30-2008 10:09 AM

Re: Guardian OS units
 
I believe it used the std 1din 250W server PS. You will need to search the PN. Also Dell is a good source for parts, besides eBay.

yfpang 07-31-2008 11:42 PM

Re: Guardian OS units
 
hi blue68f100,

Thanks for your advice as for our final resort we manage to borrow 1 working hdd from our friend to boot up our unit. And once it's up we add back our hdd into the unit and it manage to do a boot sync.

And once we have manage to boot sync 1 hdd we shutdown the unit and take out the new hdd and boot via our old hdd, and re-sync the rest of the hdd. and currently it's working fine already.

And for this do you think it's possible to copy the sector 0 to 10k to a CD rom for keeping, as the unit can be booted up from it's usb storage. As this will be easier to recover from a corrupted OS.

Thanks blue68f100 again for your help

blue68f100 08-01-2008 09:30 AM

Re: Guardian OS units
 
The Guardian OS reserves 10gig for it's shell (Raid1 across all drives). So more than 10k will be required. I use a 20-40 gig hd as a backup, just encase the unit will not boot. As long as you do a cold boot it will only copy the OS. Some use an identical HD as the raid
, this way they have a BU HD ready.

chrisnz 10-28-2008 06:23 PM

Re: Guardian OS units
 
SNAP 4500 with guardian os 4.5 and expansion shelf.

xponet 12-22-2008 07:08 AM

Re: Guardian OS units
 
I have a 4400 running version 5.0.133 SP1 with 4x320 in RAID5

hyperjason 02-17-2009 05:50 PM

Re: Guardian OS units
 
I have 3 - 1TB Snap 520's running OS 4.2.054

schnappi 03-19-2009 11:14 AM

Re: Guardian OS units
 
I have a lot off Snap Server 4400, 4200, 4500, 110, 410, 520 and many, many, many Problems with the new Guardian OS 5.1.041 - (If I install V5.1x the 4xxx Snap Servers get very, very slow).

eschw95458 03-29-2009 11:55 AM

Re: Guardian OS units
 
How much memory are you running in those units. I noticed that snap server 110 originally came with 256mb mem and they quietly upgraded them to 512mb. I would suggest upping the memory in those units and see if that helps.

blue68f100 03-29-2009 06:17 PM

Re: Guardian OS units
 
I would consider 1 gig as min. If you have a lot of users max it out.

blue68f100 01-02-2010 01:27 PM

Re: Guardian OS units
 
I keep a (2) 20gig HD loaded with just the GOS, clean install for emergencies. When I first got my unit I did intensive testing to see how well it handled crashes during read and write process involving loosing power. I was very impressed that if you had snapshots and the required settings back upped it always recovered on it's own. I did buy a APC SmartUps 1500 with the network card for my server. It has enough power to carry it and 1 PC for about 90min. When it gets to low power it shutsdown my servers.

Your better off if you can setup a USB memory stick to boot off of. 1 user did that, but due to parts of the GOS software he could not distribute it.

Like I said 1 gig is a min for GOS units. Andy says 2 gig will give the best performance. But it all gets down to how many users are hitting your server.

The SnapOS units do not benifit much unless you have large HD's and running JAVA.

Terry Kennedy 01-04-2010 06:12 AM

Re: Guardian OS units
 
Quote:

Originally Posted by blue68f100 (Post 172162)
The SnapOS units do not benifit much unless you have large HD's and running JAVA.

Another use for more memory in the SnapOS units is to avoid falling back to swap-based fsck after an unclean shutdown. I prefer to pro-actively install the maximum memory, as the last thing you want to do is suffer through a 12-hour-plus fsck in this case, and you certainly don't want to install a larger memory module when the server is in that unclean state.

This probably isn't important for 4100's with smaller drives, but when you get up to the 120GB maximum drive size, it makes a huge difference.


All times are GMT -5. The time now is 03:14 AM.

Powered by vBulletin® Version 3.7.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
(C) 2005 ProCooling.com
If we in some way offend you, insult you or your people, screw your mom, beat up your dad, or poop on your porch... we're sorry... we were probably really drunk...
Oh and dont steal our content bitches! Don't give us a reason to pee in your open car window this summer...