Bad Power Shut Down Now No Admin Access

[teastman]

Okay - this is screwy and I didn't do it. Former business owner accidentally unplugged the Snap 410 from it's UPS and it seems to have forgotten a lot of things.

I can ping it over the network at least it remembers it's static IP.

Snap Sever Manager sees it as online (although it's not licensed to manage through SSM)

Can't log in via the web interface, It doesn't remember who the users are. I even tried to do the "Limited Reset" using the reset pin hole in the front of the unit.

That "limited reset" would not reboot the server believe it or not. I was hoping that would reset the admin password but it did not.

I am trying to login through a Putty terminal to do a syswrapper to send to "tech support" but even there I am not allowed to log in. I used 115200, 8, n ,1 ,no flow control. This gets me to the login prompt on the command line but still no admin access.

I have a SnapDrImage file I could reload but now I can't login to even get a Recovery Console.


You guys have any ideas here?

Edit for machine particulars;

Server Name *******
Server Number *******
Server Type GuardianOS
Model 410
OS Version 4.4.049 SP3
BIOS Version WAV30403
Service Pack SP3
Serial Number **************
Ethernet 1:
IP Address 192.168.1.200 (DHCP) (Discovery)
Subnet Mask 255.255.255.0
Default Gateway 192.168.1.2
DNS Server 192.168.1.2
Domain Name *******.*****
WINS Server None
Ethernet (MAC) Address 00:C0:B6:22:9B:34
Ethernet 2: Ethernet has no link.
Ethernet (MAC) Address 00:C0:B6:22:9B:35
Java JVM Version 1.4.2_13-b06
Java Framework Version Unknown
Web Port Unknown
Memory Size 1024 MB
Platform Bytes 07.07.00
NIC Count 2
Drive Count 4
Power Supply Count 1
Fan Count 3
CPU Count 1
CPU Speed 1995 MHz
Expansion Unit Count 0
Discovering Server Unknown
Configuration Status Not configured
Licensed for Snap Server Manager No

[blue68f100]

If the machine was not shutdown correctly you said, it may be checking the data. This can take some time and will only mount the share once it is confirmed valid no errors.

So leave it on and see if it will eventually come up. But every cycle your just compounding the problem if it's trying to check disk.

[teastman]

Yup - I agree. However I let it sit and run all weekend and there's no change. That is when I posted the original question. It is confusing to me that the "reset" pin on the 410 does not seem to do the limited reset to default the admin password.

I have gotten a hold of Overland Tech Support - the initial helpdesk person elevated it to level 2 and I am currently waiting on a call-back.

[Max8]

The GOS Snapserver has 2x Kryptonite's

1. Stray Linux Comands - Linux people who don’t know any better, finding the CLI and trying their favorite red hat comments...
Fortunatly this is rare but when it happens they totally deny it, even after you can clearly say “it’s noted in the syswrapper!” and you are left with an IT manager who by this stage has made such a tool out of himself that he cannot do anything other than keep the tool persona going…

Once I sold a Snap 520 with 2x expansion shelves and some low level ass-clown did his little linux tweek while he was interfering with himself and it never worked right again, after the syswrapper showed us what happend, they were too embarrassed to get further help (they could have very easily reinstalled) so what does any IT manager with too much ego do? They took the HBA out of the 520 got it to work in another linux box and hooked the Expansion up to that… I wouldn’t care normally but it was my tax money that paid for that snap 520 that then sat on a shelf to collect dust when it was 4 years later when I last spoke with them…

- me after hearing about the waste

(don't care about tax money getting wasted as I am used to it!)

2. Hard Shutdowns – like Russian roulette you don’t exactly know which one will do it but 1 in 10 or 1 in 20 or less or more? will result in a corruption…
People are too used to windows locking up and if their windows system has an issue they decide to do a hard reboot… if they get the IP address wrong, or their kids failing blocks or what ever reason and the IT guy is not around they do a hard reboot, these all show up in syswrappers.

Usually once the unit locks up daily they decide to call support make up a story and when we tell them there has be 47 hard shutdowns logged in the syswrapper the story changes…

There are two ways to shutdown a snap, (1. Through the GUI, (2. A short but firm stab on the PWR button once you are sure there is nobody connected…

What is a hard shutdown? - holding in the PWR button for a few seconds is a hard shutdown, Confusing for PC users as they are used to the 4 second delay – Maybe it should be disabled so that the only hard shutdown available is pulling the PWR cord? – At least that way the user will not confuse hard shutdown with a 4 second delay in the average windows system…

[teastman]

Max8 - interesting comments. Anything to mention pertaining to the observations and questions in the original post?

[Max8]

Regrettably no, but I have faith in the techs who are assisting you, you are taking the best course of action you can.

It just seemed like an opportune time to note what I have found as the worst two things you can do to any SnapServer, apologies for not being able to assist you directly but I hope the forum can get something out of my input.

I have been on calls with some good SnapServer techs in the past and I have seen them bring lost units back from similar situations like yours, I hope the techs can do the same for you too…

I am certainly looking forward to hearing back that you had a positive outcome.

Good luck…

[Phoenix32]

Originally Posted by teastman

Max8 - interesting comments. Anything to mention pertaining to the observations and questions in the original post?

I suspect Max was just telling you in a gentle round about way what I was also thinking, sounds like your OS got corrupted maybe.

[teastman]

Originally Posted by Phoenix32

I suspect Max was just telling you in a gentle round about way what I was also thinking, sounds like your OS got corrupted maybe.

After being on the phone most of the week with Overland tech support we've come to the same conclusion. I am very fortunate that I am able (finally) to extract project data off this thing. I am doing so as of this moment.

After that I will do a complete re-install which of course whacks all data but this will show us on the next powerdown & restart whether it's a deep corruption or an actual hardware issue.

Earlier this week we had this thing up and talking and allowing users access and it all looked good, even with repeated (intentional) restarts. I shut it down through the console - moved it back to the server room - started it back up and it came right up to the Guardian Recovery Console with the amber and green alternating system LED light. Not good.

So that is where we're at. Get the data off, then nuke & pave it with OS reinstall and SP3 to see if we can actually trust this thing again.

Thanks you guys.