Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie
Hi there,
There is an issue with role permissions that is being worked on at the moment.
If you are having trouble with access or permissions on regional forums please post here to get access: https://www.boards.ie/discussion/2058365403/you-do-not-have-permission-for-that#latest

Home server problem

  • 12-11-2014 4:56pm
    #1
    Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭


    I'd appreciate anyone who's more familiar with WHS (2012 R2) or has the C2750D41 to chime in with this one.

    Machine is hanging randomly. Usually a couple of times a day, during heavy read/write to the hard drives. It seems to be only with my stand-alone drives; everything in the RAID is still fine and readable.

    The transfers will just sit where they are, and I won't be able to access much of anything on either of the drives. I've pulled each one and tried with them separately, and each time the machine has still hung, so it's not the drives, but I'd rather not have to RMA the board...

    Oh, and I've also updated chipset drives as well, no help. Currently installing a couple of Windows updates.

    Any ideas?


Comments

  • Registered Users, Registered Users 2 Posts: 5,607 ✭✭✭Nollog


    I've been using 2012 r2 dc version for about 2 months now and set up a storage pool with 5 (currently 6 because i've been swapping around disks as i bought a few new ones), the only unexpected issues i've had have been my power going out twice.

    I've got a hp microserver.
    Would you be up for trying a storage pool, they're pretty easy to set up.


  • Registered Users, Registered Users 2 Posts: 2,318 ✭✭✭deceit


    Is there are errors in event viewer in application or system?


  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    There are actually. A bunch of "storahci" warnings in System.

    0x80000000000000 Reset to device, \Device\RaidPort2, was issued.

    As for a storage pool, I'm already running FlexRAID, which is something I'm familiar with, and trust. I've heard too many horror stories about Windows' Storage Spaces to trust something like it with over 15TB of data, to be blunt.


  • Registered Users, Registered Users 2 Posts: 36,170 ✭✭✭✭ED E


    As you know I've got a very similar rig (and it seems linus will too soon), no such problems. Scheduled a pause on some services when flexraid is updating and no more failed validates.

    How many of the sata ports have you got loaded up atm? I'm only using 4 so my sata controllers won't be as busy.

    Really seems to call for the IPMI, set it to record the 60s before a crash and leave the task manager/resource manager open on the desktop all the time so you'll have a view of it in its "death throws"?


  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    I've got seven of them used at the moment. One of the Intel 6Gb/s ones (OS drive), and I think all of the Marvell ones.

    I can't predict the crashes like that. They usually take an hour or so to kick in when I start a big write going.


  • Advertisement
  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    Ok, so it's nothing to do with the hard drives. I left the machine idling, with the FlexRAID array stopped, and all maintenance tasks I could find disabled. Absolutely going. Then around 11.00 last night it starts throwing that storahci error again continually. Thinking faulty SATA controller.


  • Registered Users, Registered Users 2 Posts: 2,318 ✭✭✭deceit


    Serephucus wrote: »
    There are actually. A bunch of "storahci" warnings in System.

    0x80000000000000 Reset to device, \Device\RaidPort2, was issued.

    As for a storage pool, I'm already running FlexRAID, which is something I'm familiar with, and trust. I've heard too many horror stories about Windows' Storage Spaces to trust something like it with over 15TB of data, to be blunt.
    Can you print screen all the errors, critical and warnings you get and I will see if I can find something from them?


  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    Can do.

    89jExfnl.png

    There's the storahci error. I've updated the drivers for the Marvell controller, going to see if that helps anything before I change anything else. Hopefully you can shed some light on that.


  • Registered Users, Registered Users 2 Posts: 2,318 ✭✭✭deceit


    Do you have Aggressive link power management (ALPE) and/or Aggressive Slumber / Partial Management (ASP) features enabled in your bios?
    Edit:
    Also did you enable ahci before you installed the OS?


  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    I've no clue, to be honest. I'll check once a couple of these tests are done. Some sort of power management I take it?


  • Advertisement
  • Registered Users, Registered Users 2 Posts: 36,170 ✭✭✭✭ED E


    Google says thats even ID 129. Searched my logs. Got nada.

    Your diagnosis of a dud controller could well be it. If I were you I'd boot another OS off a stick for 48hrs and just have it idle. See if she bites the bullet or not?

    EDIT: Didnt refresh the thread, hadnt seen your posts.


    Also, slightly on topic:


  • Registered Users, Registered Users 2 Posts: 2,318 ✭✭✭deceit


    Serephucus wrote: »
    I've no clue, to be honest. I'll check once a couple of these tests are done. Some sort of power management I take it?
    Yes it is and with alot of boards can cause a lot of freezing issues with similar errors. I've seen it on proper server boards but its possible workstation boards have the features.


  • Registered Users, Registered Users 2 Posts: 2,318 ✭✭✭deceit


    I just looked through the manual for it:
    Sata Aggressive Link Power Management is a feature of the board, ensure this is disabled. This will most likely cause issues when it kicks in with multiple disks.
    Usually high end raid cards force this feature to wait until they are done before enabling where as a lot of software raid options and cheaper raid cards don't pay attention to it which is why you disable it.
    Hopefully disabling this will help


  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    Cool, thanks. Assuming it locks up again that's the first thing I'll change. Will probably change it the next time I restart anyway.

    Edit: Restarted there, and SATA Link Management was disabled on all the controllers.


    Edit2: About half an hour into a bunch of writes, and most of the drives (I'd assume the ones on the Marvell controllers) just dropped out of the machine. Me-thinks I might have a faulty controller alright...


  • Registered Users, Registered Users 2 Posts: 5,607 ✭✭✭Nollog


    Serephucus wrote: »
    There are actually. A bunch of "storahci" warnings in System.

    0x80000000000000 Reset to device, \Device\RaidPort2, was issued.

    As for a storage pool, I'm already running FlexRAID, which is something I'm familiar with, and trust. I've heard too many horror stories about Windows' Storage Spaces to trust something like it with over 15TB of data, to be blunt.

    It's mainly media, but I did a lot of research on it, and seemed great for easy adding drives and raidlike performance, curious what issues people have, there's great help online about any issues, I had some trouble with the GUI so looked up power shell commands to do the stuff I needed.


  • Registered Users, Registered Users 2 Posts: 36,170 ✭✭✭✭ED E


    OT but have you tried to Hot Swap Seph? Did it work?


  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    I haven't, but it should work fine. The SATA protocol is actually hot swap capable anyway (as long as you don't have any of those performance enhancements in Device Manager ticked)


  • Registered Users, Registered Users 2 Posts: 36,170 ✭✭✭✭ED E


    Serephucus wrote: »
    I haven't, but it should work fine. The SATA protocol is actually hot swap capable anyway (as long as you don't have any of those performance enhancements in Device Manager ticked)

    Turns out it was bay #8 not working, not the hot swap feature. Think the screws from the base of the drive cage were shorting the pcb! Gonna have to replace them with shorter ones.


  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    Interesting. I'll give that a look the next time I shut down the server. Haven't tried the bottom bay yet.

    Have the board in for RMA. Being collected on Tuesday, hopefully will get a new one sent out by the end of the week.


  • Registered Users, Registered Users 2 Posts: 36,170 ✭✭✭✭ED E


    Interested to see what their RMA process is like, though fingers crossed I never need it.

    Pain in the face for you. Fingers crossed it goes well.


  • Advertisement
  • Registered Users, Registered Users 2 Posts: 36,170 ✭✭✭✭ED E


    *Bump!*

    Firstly, how did the RMA end up?

    Secondly, are you using an any offsite backups? Was going to go with Backblaze but no server '08 :o Thinking crashplan now, but open to anyone with trust no one.


  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    Completely forgot about this, my bad.

    Well, I had sent the board off to Scan, and two days after it arrived at their warehouse, they got stock of the board back, but because there was no stock when my board arrived, it got sent to the Netherlands for repair, so I had to wait an extra two weeks (in theory).

    The board arrives back at Scan, and for some reason no-one thought to send it out to me, until I called a week later. All in all, just over a month with the server down. Not the best RMA ever.

    Got the board back in, only to find that same hard drive was still dropping out. Turns out it was dying as well, so I had to RMA that.

    WD's RMA is far better, thankfully. Went for the Advance RMA, so I had a replacement drive within five days. I'd like to point out that this includes a weekend, and New Years, so I was pretty impressed. Everything back up and working nicely at the moment.

    As for off-site backups, I don't use anything. The only stuff that's important enough is just documents and pictures, and they're on BT Sync.


  • Registered Users, Registered Users 2 Posts: 36,170 ✭✭✭✭ED E


    Poor show scan. If only everybody could be like amazon for CS. Fingers crossed though I dont ever need to go near the board again, except maybe to add RAM.


    I have some stuff in gDrive, some in Dropbox and a tiny bit in Tresorit. But all my raws from the DSLR and some footage etc is currently "fire/flood vulnerable". Statistically thats not a huge risk, but because of how much I preach good backups to all the non-nerds in my life I really cant continue without 100% offsite replication.


  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    Agreed. I've only had to deal with Amazon once or twice, but it's always been completely painless.

    Time to buy another 380 and D4I so! :P


  • Registered Users, Registered Users 2 Posts: 36,170 ✭✭✭✭ED E


    Have to populate this one fully first! Gonna have camelcamelcamel alert me if the 2TB reds go on sale, need more storage e-peen :pac:


  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    The 4TB drives are cheaper per GB.

    http://geizhals.de/western-digital-wd-red-4tb-wd40efrx-a992027.html

    €80/2TB vs. 90.

    How much storage have you got in there anyway?


  • Registered Users, Registered Users 2 Posts: 36,170 ✭✭✭✭ED E


    3x2TB, 1s the PPU and 2x DRUs. Plus a 160GB for gDrive as it refuses to run on the array and as a backup destination for the SSD so the OS/Flexraid config is safe in case of an SSD death.

    2TBs were the sweet spot when I ordered, 4TB makes much more sense in terms of bay consumption. I could buy 2x4's and migrate the existing PPU to a DRU. So 10TBs in DRUs and 4TB in PPU. Its a start anyways.


  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    Nice. Actually, what are you using for the OS drive backup? I'm using the built-in backup yoke in Server 2012, but I'm not a fan of it.

    If only there was a bulkrate price for hard drives...


  • Registered Users, Registered Users 2 Posts: 36,170 ✭✭✭✭ED E


    Same, "Windows Server Backup" in bare metal mode with VSS. I wanted to use Acronis as have a free licence from an SSD, but no dice with a server OS. As far as I know it should be a straight restore to the new drive job if the SSD ever bit the bullet, but I must verify that fact.


  • Advertisement
  • Registered Users, Registered Users 2 Posts: 7,181 ✭✭✭Serephucus


    I think I have a copy of Acronis as well somewhere actually, might try that out. I remember using it a year or two ago, and I wasn't hugely impressed. I'd imagine it's changed by now, but at the time it didn't handle backups correctly. I had it set to just do a complete backup once a week, but it never deleted the previous ones. :P


  • Registered Users, Registered Users 2 Posts: 36,170 ✭✭✭✭ED E


    My main machine uses Norton Ghost which came with the samsung 830, that works really well. Was just trying acronis because I had it.


Advertisement