jumperalex

Members
  • Posts

    2000
  • Joined

  • Last visited

Posts posted by jumperalex

  1. So here is what happened ...

     

    Short version, I'm an idiot who was being lazy.

     

    Long version, I had already shut down my PC before shutting down my unraid server.  So I fired up the console, and typed "shutdown -now".  Yeah problem right.  I realized it based on the incomplete powerdown so I reboot with the power button to find a parity check in progress.  Ok fine, but I still want the server shutdown during my vacation, I'll run the parity check later.  Problem is, now I can't get into the GUI or Unmenu.

     

    I get back from vacation and I have the same problem.  Boot, get into gui/unmenu, see parity check in progress, few minutes later I can't get into the gui, or ssh, but I can still hit the console.  I try to restart unmenu via "uu", but no joy.

     

    Reboot again, quickly get into the gui, stop the parity check, and bam, things are good for half an hour. So I do a clean reboot, wait another 30 minutes, reboot, and then manually start a parity check.  Now 30 minutes of parity check going and I can still access the server.  All seems happy happy joy joy ...

     

    So why are we here?  Well with all that going on I am now looking very closely at the syslog to see if there are lingering problems.  Only thing is, I know there are "errors" we don't need to worry about but I can't separate the wheat from the chaff.  I have attached my current running syslog and the syslog saved after my last shutdown. 

     

    I just want to know if there is anything still wrong.  Any help is greatly appreciated.

    My_two_syslogs.zip

  2. So you've pulled all add-on cards right?

     

    And you've installed a known good video card?  PCI-E? ... can you try an AGP or PCI card in there?

     

    Debug codes are nice but ...

    Have you swapped in known good memory?

     

    Have you swapped in a known good power supply?

     

    Beyond those you might try a CMOS reset in the hopes that your bios got scrambled and you can get it back. 

     

    Otherwise, your only left with getting a new motherboard.  At least we survive on cheap components here at Unraid ;-)

  3. Ben

    probably because the buffer size used by the driver for the AOC-SASLP-MV8 is half the size of that used by the driver for the chipset on the motheboard.  It therefore needs to read two "buffers" worth to access the same amount of data as the other disk controller gets in a single read. 

     

    So you are saying:

    1byte read operation = 1 read = 2byte read operation

     

    as opposed to

     

    1byte read = 1 read

    2bytes read = 2 reads

  4. Just wanted to mention the upcoming NZXT Source 210.  It's supposed to be very affordable (~$40) and with 3 5.25'' bays and 8 3.5'' internal bays you can run a single 5 in 3 cage and hit 13 drives for significantly less than the cost of running a 15-bay tower build.  It's also very slick in terms of aesthetics.

     

    I'm about 10 seconds away from pulling the trigger on this very case.  I almost bought the Antec 100 with 6 internals, but decided to search some more.  I frankly see no need in my world for 5x3 bays since they just cost money, add noise, and can't possibly be as cool running as the slightly more spaced out 8 drive stack in the NZXT.  And for the price, sheesh how can I go wrong :o $39.99 + $9.99 shipping from newegg.  Not so sound like a shill but here is a link to some sweet pics on the NZXT web site http://www.nzxt.com/new/products/classic_series/source_210 ... can anyone tell me .. .it looks like they have the slots for tool-less install but not the actual guts to make it happen.  Am I mistaken?  Can I buy those somewhere if I'm right?

  5. If you're going through all the trouble - maybe it's time to consider some off site backup?

     

    I'm still looking for the "easy to use" UnRaid to UnRaid Offsite solution.  :-)

     

    Russell

     

    As I said, this isn't meant to be a replacement for off-line or off-site backup.  Just another layer of protection.  I've got the HD space, I've got the drive, and I've got the paranoia ;-)  I already backup to an external HD from my PC, and to Unraid from my PC.  But I'd like to make an unraid double drive failure less likely to take out my most recent data.  Unraid takes daily snap shots, the external is about weekly ("about" because it is not physically connected or powered to prevent electrical surge concerns so I have to do it manually).

     

    For off-site I'm most likely just going to set up a sync to my FTP server hosted in Utah.  I'm in VA, so I figure I have enough geographical separation.  If both sites are taken down ... well ... let's just say I figure I'll have more important things to worry about than if I lost those pictures from that drunken night in bangkok >;-)

  6. STFF  ;)

     

    Ideas like this have been discussed here many times in the past. Basically, set up cron to run rsync on a regular basis or something similar to that.

     

    Peter

     

     

    I swear I searched at work.  Then I get home, read the two responses, try another search and find the exact same question asked in March [sigh] ... thanks for indulging me guys.  Still seems like this is a "feature" worth adding because right now I'm still struggling to figure out rsync, cron, crontab, and how to make it work through reboots. :\  But I like to learn, so I have no problem searching more and reading man pages.  Cheers.

  7. So I just finished reading a nice little fanboy war over at XBMC about freenas vs unraid and it got me thinking ... for our standard media files 1 parity worth of protection is enough (otherwise we wouldn't be using unraid), but what if we wanted more for a "small" subset of our data?  "Small" in this case is relative in so much as family photos, word documents, email backups, etc are likely orders of magnitude smaller than the terabytes of video and music files many of us have.

     

    My thought was, why not keep multiple copies of that important data on the unraid box ... like ... say ... a copy on every single drive in the system, or half, or even just three or four?  That way even losing two, three, or more drives doesn't mean the loss of that important data.

     

    Don't get me wrong, for the really paranoid this isn't a replacement for off-line and/or off-site backups, but it seems like a reasonable strategy to add another layer or two of redundancy for some data we might really like protected; especially as a bridge between backups.

     

    Sooo all that said, is there an easy way to do this in unraid?  Set up a user share "My Important Data" with split level 0 and then have it replicated across multiple drives?  I figure a script could be used to periodically rsynch (?) between “My Important Data” and unexported shares created on the other disks? I'm only a duffer but I’d be willing to give it a shot.

     

    Ideally this might someday be added as a feature.  Perhaps when you create the share there is also a replication check box with associated drive selection drop-down.  Then unraid would be able to write the duplicates in real time, or with some reasonable user selectable delay.

     

    I’m purposely avoiding using the term mirror here since in my head that applies to entire drives and not subsets of data.

     

    Thoughts?  Suggestions?  Pitch Forks?  RTFM’s? :)

     

  8. I don't this exact m/b but I do have the 8112L. I'm also fairly new (about 15 total days uptime and 3 days continuous max uptime), but I've sort of tried to stress my new box with four "failures" that included four parity swaps (install replacement drive larger than old parity) and moving ~1TB of data over the NIC, and so far so good.  With nothing that looks to be a problem, or a NIC slow down.

  9. It is re-created from the /boot/packages/ssmtp package.auto_install script every  time it re-installs on reboot. 

     

    Well then I assume the better, and easier, solution would be to edit the install script from unMENU?  Certainly easier than a manually edit of ssmtp.conf from the console and add a cp line to the "go" script.  I just did that now, but have not rebooted because I can't take the array off line at the moment.  Is there any reason not to do that?

  10. Ah sweet.  Thanks.  I was just about to ask if there was a way to do that.  I figured there was, but I didn't know where, on flash, to go about doing that.

     

    Slightly OT, but it is my thread after all ;) ... where is ssmtp.conf stored on shutdown?  Is it in bzroot?  Given your proposed "patch" I'm assuming it isn't anywhere that I can actually edit rather than using your method.

  11. :'( no go.  when I tried adding "\" the config screen now shows the password correctly, however, what is actually stored in ssmtp.conf is still truncated and I get an Auth Fail.  Manual edit again solves the issue.

     

    any other thoughts?  I suppose might be considered a bit of a "bug" worth putting on your list of things to do and/or adding a note to any instructions about it.

     

    For the record, thanks so much for all your work with unRAID and unMENU, I'm a new user and all the info here has made it so easy to set up my first server that this is the first question I've had to ask.

  12. System:

    unRaid 4.7

    unMenu 1.4 Revision: 223

    ssmtp_2.62.orig.tar.gz ...

     

    Problem: Auth Failure using gmail.

     

    So I narrowed the problem to the fact that my password contains a $ in it.  Whenever I add my password and press save, all is good (read: the screen still displays my full password).  Then i click reinstall with new settings and the password displayed is now truncated starting at the $ (abcd$efg => abcd).  As a test I played around with the location of the $ and sure enough, the truncation point moved with it. 

     

    Solution: Then I went in and modified the .conf file in /usr/local/etc/ssmtp with mcedit.  The package manager still shows my truncated password after a screen refresh, but guess what, unRaid can now send mail with no trouble.

     

    This brings up two questions:

     

    Q: Will my manually edited .conf file survive a reboot?

    Q: Is there a permanent solution to this behavior?

     

    Thanks

    Alex

  13. Unfortunatly I bought my M4A785-M when I was only considering FreeNAS so Im' stuck with it until I spend the money for a new MB.  That said I also am not in a position to need more than 6 drives anyway, and even 8 or 10 using two PCIE SATA cards is a long way off.  That said, as a noob to unRAID I'm hoovering the forums for info.  This thread just makes me wonder if there has been a motherboard bios upgrade that solves the problem.

     

    My board is fairly new so I'm on the current 1006 bios.  Raja, based on your build date it is "possible" that you were on 0906 since the release date for 1006 was 12 Nov 2010 and you might have gotten boards that had been sitting on the shelves with the older BIOS.  I know, not really that likely, but it is possible [shrug] :)

     

    Any chance you noted the bios version during that build?

    [EDIT] sigh ... ok rereading your "doozey" thread I see that you DID upgrade the MB bios (hey I was right in that one respect  ;)  )

     

    As for flashdrive booting, I've booted this MB with three completely different (and from different time periods) flash drives with no trouble (yeah I know, that doesn't mean you didn't have problems :o ).  The only thing I noticed is that I always had to go into the BIOS to tell it to use the "new " flashdrive as the primary boot device and then from there on in it boots fine.  So my point of saying that is, where ahve all the reoprts been coming from about it being finicky about what flash drive it is willing to boot from?