SSD Posted June 24, 2011 Share Posted June 24, 2011 Thanks, I never knew about Screen. Just to confirm my reading of the Screen documentation: if I install Screen and run preclear via Screen, I CAN quit the Screen telnet session and the preclear will still be running. I can also then shut down the PC and re-attach to Screen (and the still running preclear session) after I restart the PC. Is my understanding correct? You can't "quit" screen, but can "detach" from the screen session by pressing [Crtl]A followed by the "d" key. You can then resume screen with the "screen -r" command. Quote Link to comment
sonofdbn Posted June 24, 2011 Share Posted June 24, 2011 Thanks for the clarification. My concern is around what happens in between detach and resume. Can I shut down the PC in between without affecting the preclear? Quote Link to comment
SSD Posted June 25, 2011 Share Posted June 25, 2011 Thanks for the clarification. My concern is around what happens in between detach and resume. Can I shut down the PC in between without affecting the preclear? Yes - you can resume the screen session from any telnet session (from another machine, from the unRAID servers, etc.) Quote Link to comment
sonofdbn Posted June 25, 2011 Share Posted June 25, 2011 Many thanks for all the help. Quote Link to comment
CrashnBrn Posted June 25, 2011 Share Posted June 25, 2011 2 quick questions:) Are the only Advance format drives WD? I have two 2TB Hitachi drives and I'm assuming they are NOT Advance format drives, is this correct? Also how can I preclear 2 disks at once? Many thanks! Edit: Running 4.6 Quote Link to comment
SSD Posted June 25, 2011 Share Posted June 25, 2011 From inside screen, press [ctrl]a followed by "c". This will create another telnet session within screen, where you can run another preclear. Press [ctrl]a followed by space key to cycle through all sessions within the screen process. Quote Link to comment
SSD Posted June 25, 2011 Share Posted June 25, 2011 I do not believe any of the Hitachi drives are AF, but it should not matter. Treat all drives the same. Use the -A option to preclear them. Quote Link to comment
Joe L. Posted June 25, 2011 Share Posted June 25, 2011 2 quick questions:) Are the only Advance format drives WD? I have two 2TB Hitachi drives and I'm assuming they are NOT Advance format drives, is this correct? true, BUT any drive can be addressed with a partition that starts at a 4096 byte boundary. Forget the words "advanced format" as it is mostly marketing as far as the real usefullness to us. Internally on the disk it means that EVENTUALLY they can provide more data, or better data reliability. So far, as I said, it is mostly marketing for the 2TB disk size. The 2TB Advanced format drives are exactly the same size, and basically the same reliability as non-advanced-format. Older drives can only be accessed on 512 byte boundaries... but guess what, every 4096 byte boundary as ALSO a 512 byte boundary. unRAID (and all window's OS) historically started their partitions on sector 63. This was 63 * 512 bytes on the disk, as disk sectors are (historically) 512 bytes on all disks manufactured to date, even the WD Advanced-Sector disks. Now, unRAID can start a disk partition on sector 64, and 64 * 512 is also a multiple of 4096. Any disk will work at its optimum speed. Lastly, unless you are in a laboratory, it really does not make much difference if you are using your unRAID server for music, of movies, or pictures. If you are watching a movie, for the most part it is stored sequentially on the disk, and if the disk platter has to make one additional rotation to get the very first bit of a movie, then it will take 1/5400th of a minute more to return the result since the disk platter must make one more revolution. Somehow, unless you are in a lab running tests, you would not know the difference. Also how can I preclear 2 disks at once? To preclear 2 disks at one, use two telnet sessions, two system consoles (there are 6), or install and use "screen" Joe L. Quote Link to comment
CrashnBrn Posted June 25, 2011 Share Posted June 25, 2011 I do not believe any of the Hitachi drives are AF, but it should not matter. Treat all drives the same. Use the -A option to preclear them. Thanks for the replies Joe and bjp999, always wonderful information! To the above quote, since I'm running 4.6 and have not used the -A command on any of my drives, will this be an issue? If on version 4.7 of unRAID and you are clearing an ADVANCED FORMAT drive that works best with a 4k alignment of its data, use the "-A" option as in preclear_disk.sh -A /dev/sda This will result in a starting sector of 64 for the resulting partition that will be used for unRAID's file system. This is NOT backwards compatible with any version of unRAID prior to 4.7. Quote Link to comment
Joe L. Posted June 25, 2011 Share Posted June 25, 2011 I do not believe any of the Hitachi drives are AF, but it should not matter. Treat all drives the same. Use the -A option to preclear them. Thanks for the replies Joe and bjp999, always wonderful information! To the above quote, since I'm running 4.6 and have not used the -A command on any of my drives, will this be an issue? If on version 4.7 of unRAID and you are clearing an ADVANCED FORMAT drive that works best with a 4k alignment of its data, use the "-A" option as in preclear_disk.sh -A /dev/sda This will result in a starting sector of 64 for the resulting partition that will be used for unRAID's file system. This is NOT backwards compatible with any version of unRAID prior to 4.7. good, because if you had, they would not have been recognized as pre-cleared in any release of unRAID prior to 4.7. No harm would occur, but unRAID would write zeros to them for you again if you added them to an array with established parity. Joe L. Quote Link to comment
SSD Posted June 25, 2011 Share Posted June 25, 2011 To the above quote, since I'm running 4.6 and have not used the -A command on any of my drives, will this be an issue? If on version 4.7 of unRAID and you are clearing an ADVANCED FORMAT drive that works best with a 4k alignment of its data, use the "-A" option as in preclear_disk.sh -A /dev/sda This will result in a starting sector of 64 for the resulting partition that will be used for unRAID's file system. This is NOT backwards compatible with any version of unRAID prior to 4.7. Saw the reference to 4.6 but didn't register. I'd recommend you move to 4.7, and set the option for aligned partitions. Takes all the guess work out of it. Preclear is even smart enough to know to preclear the disks based on that setting. Quote Link to comment
CrashnBrn Posted June 25, 2011 Share Posted June 25, 2011 To the above quote, since I'm running 4.6 and have not used the -A command on any of my drives, will this be an issue? If on version 4.7 of unRAID and you are clearing an ADVANCED FORMAT drive that works best with a 4k alignment of its data, use the "-A" option as in preclear_disk.sh -A /dev/sda This will result in a starting sector of 64 for the resulting partition that will be used for unRAID's file system. This is NOT backwards compatible with any version of unRAID prior to 4.7. Saw the reference to 4.6 but didn't register. I'd recommend you move to 4.7, and set the option for aligned partitions. Takes all the guess work out of it. Preclear is even smart enough to know to preclear the disks based on that setting. As far as two drives at the same time, is it possible to do it directly from console? I don't have a network connection to the machine I am using to preclear. Quote Link to comment
Joe L. Posted June 25, 2011 Share Posted June 25, 2011 To the above quote, since I'm running 4.6 and have not used the -A command on any of my drives, will this be an issue? If on version 4.7 of unRAID and you are clearing an ADVANCED FORMAT drive that works best with a 4k alignment of its data, use the "-A" option as in preclear_disk.sh -A /dev/sda This will result in a starting sector of 64 for the resulting partition that will be used for unRAID's file system. This is NOT backwards compatible with any version of unRAID prior to 4.7. Saw the reference to 4.6 but didn't register. I'd recommend you move to 4.7, and set the option for aligned partitions. Takes all the guess work out of it. Preclear is even smart enough to know to preclear the disks based on that setting. As far as two drives at the same time, is it possible to do it directly from console? I don't have a network connection to the machine I am using to preclear. Yes, and I already described that There are 6 virtual consoles. (Alt-F1 through Alt-F6) In Linux, the user switches between them with the key combination Alt plus a function key – for example Alt+F1 to access the virtual console number 1. Alt+Left arrow changes to the previous virtual console and Alt+Right arrow to the next virtual console. When you first boot unRAID, you are in virtual console 1. Use Alt-F2 to get to a second virtual console, Alt-arrows, or Alt-F1 to get back to the first. You can get 6 disks pre-clearing at the same time from the console. Or, as I also said, install "screen" and invoke it on a single virtual console, with virtual "screens" and have as many as you like. Joe L. Quote Link to comment
CrashnBrn Posted June 25, 2011 Share Posted June 25, 2011 To the above quote, since I'm running 4.6 and have not used the -A command on any of my drives, will this be an issue? If on version 4.7 of unRAID and you are clearing an ADVANCED FORMAT drive that works best with a 4k alignment of its data, use the "-A" option as in preclear_disk.sh -A /dev/sda This will result in a starting sector of 64 for the resulting partition that will be used for unRAID's file system. This is NOT backwards compatible with any version of unRAID prior to 4.7. Saw the reference to 4.6 but didn't register. I'd recommend you move to 4.7, and set the option for aligned partitions. Takes all the guess work out of it. Preclear is even smart enough to know to preclear the disks based on that setting. As far as two drives at the same time, is it possible to do it directly from console? I don't have a network connection to the machine I am using to preclear. Yes, and I already described that There are 6 virtual consoles. (Alt-F1 through Alt-F6) In Linux, the user switches between them with the key combination Alt plus a function key – for example Alt+F1 to access the virtual console number 1. Alt+Left arrow changes to the previous virtual console and Alt+Right arrow to the next virtual console. When you first boot unRAID, you are in virtual console 1. Use Alt-F2 to get to a second virtual console, Alt-arrows, or Alt-F1 to get back to the first. You can get 6 disks pre-clearing at the same time from the console. Or, as I also said, install "screen" and invoke it on a single virtual console, with virtual "screens" and have as many as you like. Joe L. Totally missed that sorry! Pre-clearing now! Thanks again for all of the help. Quote Link to comment
CrashnBrn Posted July 6, 2011 Share Posted July 6, 2011 So pre-cleard drive has been added, but in the normal unraid it shows there is an unformated disk present and asks if I want to format it. In MyMain (Unmenu) it shows that it's good to go. Shouldn't it already be formatted? And am I ok to format it? Quote Link to comment
Joe L. Posted July 6, 2011 Share Posted July 6, 2011 So pre-cleard drive has been added, but in the normal unraid it shows there is an unformated disk present and asks if I want to format it. In MyMain (Unmenu) it shows that it's good to go. Shouldn't it already be formatted? No, clearing it writes zeros to the entire drive and partially partitions it. It does not format it. And am I ok to format it? Yes, that is what you need to do. (as long as it is the only disk showing as un-formatted) myMain is confusing you. If the disk is not formatted, it cannot be mounted or used. Ignore it for now. Press the "Format" button. Quote Link to comment
toby9999 Posted July 10, 2011 Share Posted July 10, 2011 Hi I precleared 6 drives (sda, sdc-sdg). Then I installed latest unmenu/screen to do the rest (another 16 drives). In MyMenu, sda didn't show as a precleared drive (it's icon = "+") and it's number of writes showed quite a different value to the other 5 precleared drives. I then used screen to kick off preclearing sda again as well as the remaining 16 drives (sdh-sdw). I refreshed the MyMenu page and it showed every drive EXCEPT SDA as a precleared drive, and the usage column showed the progress of the preclear. I then noticed the last line of the syslog saying ... Tower unmenu[18653]: Precleared disk sda not found in drivedb[] (Drive related) 1) how do I fix this problem? is there a way to manually add sda to the drivedb? 2) since sda has started the preclear again (pre-read, step 1), is it safe to cancel the preclear and will it retain the precleared signature ready for when I add it to the unraid array? Thx! Cheers Toby Quote Link to comment
Joe L. Posted July 10, 2011 Share Posted July 10, 2011 Hi I precleared 6 drives (sda, sdc-sdg). Then I installed latest unmenu/screen to do the rest (another 16 drives). In MyMenu, sda didn't show as a precleared drive (it's icon = "+") and it's number of writes showed quite a different value to the other 5 precleared drives. I then used screen to kick off preclearing sda again as well as the remaining 16 drives (sdh-sdw). I refreshed the MyMenu page and it showed every drive EXCEPT SDA as a precleared drive, and the usage column showed the progress of the preclear. I then noticed the last line of the syslog saying ... Tower unmenu[18653]: Precleared disk sda not found in drivedb[] (Drive related) 1) how do I fix this problem? is there a way to manually add sda to the drivedb? 2) since sda has started the preclear again (pre-read, step 1), is it safe to cancel the preclear and will it retain the precleared signature ready for when I add it to the unraid array? Thx! Cheers Toby you are reporting a bug with myMain, not with the preclear pocess. Connect back with your screen sessions and look at the status there, cycling through each session in turn to learn what is happening. Report the drivedb[] issue to bjp999. It sounds as if his counter for referencing the drivedb array created and used by myMain is off by 1. You can test the pre-clear status of any drive by typing: preclear_disk.sh -t /dev/sdX (where sdX = the three letter device designation assigned by linux) Yes, you can interrupt the process in the pre-read phase and it should be set to pre-cleared if previously pre-cleared and not subsequently written to. Oh yes, make sure you use the most recent preclear_disk.sh program. (especially if you have a 3TB drive, as older programs will not write a valid pre-signature to those disks. A 3TB drive cannot be properly set to pre-clear status with any version of the pre-clear script < 1.12) Joe L. Quote Link to comment
toby9999 Posted July 10, 2011 Share Posted July 10, 2011 @Joe L. I was running 1.11 :-( Have downloaded 1.12beta now and will see how I go. I ran pre... -t /dev/sdc (another drive already precleared by 1.11) and it didn't find a valid signature (see attached image) - I'll obviously have to redo every drive, no problem. As the 6 drives all reported absolutely no problems (1 pass), I might use the -n switch to skip the pre/post reads on those drives. I'll contact BJP999 re: the drivedb issue, thanks. Many things are in flux at the moment with the new versions of unraid, support for 3TB drives etc. --- On another note ... I know the suggested method for running multiple preclears simultaneously is via screen, but I was just wondering if it wouldn't be easier to have that capability included in UnMenu? Or would it be too dangerous if the interface was too easy? Hmmm. Quote Link to comment
heffneil Posted July 10, 2011 Share Posted July 10, 2011 This is the output I got recently on my suspect drive while preclearing: = Post-Read in progress: 3% complete. ( of 2,000,398,934,016 bytes read ) Elapsed Time: 29:55:11 ./preclear_disk.sh: line 1080: 0+()%(243201) : syntax error: operand expected (error token is ")%(243201) ") ========================================================================1.11 == ST32000542AS 5XW1CTY2 == Disk /dev/sdi has NOT been precleared successfully == skip=3800 count=200 bs=8225280 returned instead of 00000 skip=6000 count=200 bs=8225280 returned instead of 00000 ============================================================================ Any ideas what is broken here? I am running this: preclear_disk.sh -A -c 3 /dev/sdi Thanks! Neil Quote Link to comment
Joe L. Posted July 10, 2011 Share Posted July 10, 2011 This is the output I got recently on my suspect drive while preclearing: = Post-Read in progress: 3% complete. ( of 2,000,398,934,016 bytes read ) Elapsed Time: 29:55:11 ./preclear_disk.sh: line 1080: 0+()%(243201) : syntax error: operand expected (error token is ")%(243201) ") ========================================================================1.11 == ST32000542AS 5XW1CTY2 == Disk /dev/sdi has NOT been precleared successfully == skip=3800 count=200 bs=8225280 returned instead of 00000 skip=6000 count=200 bs=8225280 returned instead of 00000 ============================================================================ Any ideas what is broken here? I am running this: preclear_disk.sh -A -c 3 /dev/sdi Thanks! Neil your disk stopped responding? It apparently did not return anything when read in the post-read. does it return anything for fdisk -l /dev/sdi or more specifically for: dd if=/dev/sdi count=200 bs=8225280 skip=3800 | sum and dd if=/dev/sdi count=200 bs=8225280 skip=6000 | sum The expected "sum" is zero. Quote Link to comment
japilch Posted August 21, 2011 Share Posted August 21, 2011 joe have a disc was preclearing when a storm hit - comp running telnet restarted - unraid server not affected still running script - am monitoring via a laptop - my main showing that it is in the last post read of 3 cycles. can I reconnect the telnet session - not showing any issues to this point . is the last set of preclear data kept somewhere - as the sever seems to have finally recogised that telnet has died. Quote Link to comment
Joe L. Posted August 22, 2011 Share Posted August 22, 2011 joe have a disc was preclearing when a storm hit - comp running telnet restarted - unraid server not affected still running script - am monitoring via a laptop - my main showing that it is in the last post read of 3 cycles. can I reconnect the telnet session - not showing any issues to this point . is the last set of preclear data kept somewhere - as the sever seems to have finally recogised that telnet has died. if running under screen, you cwn reattach. It not,it is in /tmp if you have not rebooted. Quote Link to comment
jbuszkie Posted August 22, 2011 Share Posted August 22, 2011 Joe, I can't remember... in your script between cycles.. Do you run another SMART test? I'm running two cycles on my new 3TB drive and I'd be interested to see if there were SMART differences after the first cycle. The first cycle took 45hours!! I'd hate to wait another 45 hours only to find there were errors on the first run! ========================================================================1.12 == == Disk /dev/sdd has successfully finished a preclear cycle == == Finished Cycle 1 of 2 cycles == == Using read block size = 8,225,280 Bytes == Last Cycle's Pre Read Time : 9:39:14 (86 MB/s) == Last Cycle's Zeroing time : 8:59:39 (92 MB/s) == Last Cycle's Post Read Time : 26:30:13 (31 MB/s) == Last Cycle's Total Time : 45:10:16 == == Total Elapsed Time 45:10:16 == == Disk Start Temperature: 30C == == Current Disk Temperature: 37C, == == Starting next cycle == ========================================================================1.12 I suppose I can do manual diff if it's in the /tmp area... Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.