limetech Posted February 27, 2011 Share Posted February 27, 2011 Sorry, -beta5 has been taken down & -beta6 will be up shortly to fix a seg fault bug. If you already have -beta5 downloaded, you can get it to work... look here: http://lime-technology.com/forum/index.php?topic=11372.msg108192#msg108192 Download Please read the Release Notes located on the unRAID wiki. In particular, uninstall or disable any 3rd party add-ons until the add-on author has verified correct operation with this release. Link to comment
aiden Posted February 27, 2011 Share Posted February 27, 2011 Wow, there are a lot of changes here! Especially like this little morsel: - webGui: support hot plug of hard drives when array is Stopped (limited h/w testing). Awesome work Tom! Link to comment
speeding_ant Posted February 28, 2011 Share Posted February 28, 2011 Just installed remotely from work, server isn't getting to the admin page. Will post syslog etc and revert back for now. No additions/plugins installed, just a stock standard install. Cheers Link to comment
SSD Posted February 28, 2011 Share Posted February 28, 2011 Tom - Is there anything related to large drives in here - even in alpha stage? I am assuming not but wanted to ask. Am about to create 800G HPAs on my 2 new 3T drives so I can use them until 3T support is ready, but if there is anything here I can help test before I do that, I would do that first. Thanks Link to comment
limetech Posted February 28, 2011 Author Share Posted February 28, 2011 Wow, there are a lot of changes here! Especially like this little morsel: - webGui: support hot plug of hard drives when array is Stopped (limited h/w testing). Awesome work Tom! Yes, this is pretty nice. Works so far with AHCI and SAS. Did not work with old ICH5 sata controller, but did not look into it much either; may be able to get it to work with non-AHCI. Link to comment
Zeron Posted February 28, 2011 Share Posted February 28, 2011 Just installed remotely from work, server isn't getting to the admin page. Will post syslog etc and revert back for now. No additions/plugins installed, just a stock standard install. Cheers Same problem. emhttp does not start on boot. Trying to start it manually produces a Segmentation fault. Link to comment
limetech Posted February 28, 2011 Author Share Posted February 28, 2011 Tom - Is there anything related to large drives in here - even in alpha stage? I am assuming not but wanted to ask. Am about to create 800G HPAs on my 2 new 3T drives so I can use them until 3T support is ready, but if there is anything here I can help test before I do that, I would do that first. Thanks No sorry, but more prep work was done - if you watch the boot sequence, all the drive detection, size calculations, etc., is entirely out of the driver now and passed by emhttp. Final step is to add code to create/process GPT's. Link to comment
limetech Posted February 28, 2011 Author Share Posted February 28, 2011 Just installed remotely from work, server isn't getting to the admin page. Will post syslog etc and revert back for now. No additions/plugins installed, just a stock standard install. Cheers Same problem. emhttp does not start on boot. Trying to start it manually produces a Segmentation fault. What version are you upgrading from? Does is seg fault upon first boot? Link to comment
Zeron Posted February 28, 2011 Share Posted February 28, 2011 5.0b4 Syslog here: http://pastebin.com/mBCUvBkR Link to comment
limetech Posted February 28, 2011 Author Share Posted February 28, 2011 5.0b4 Syslog here: http://pastebin.com/mBCUvBkR Thanks, looking into it. Link to comment
subwars Posted February 28, 2011 Share Posted February 28, 2011 Are there any problems changing from ide to ahci. as far as causing problems with data corruption or anything? My boards been in ide for some time now, from memory ahci used to cause problems? Remember you added a thing to disable the ncq at least anyway. Link to comment
limetech Posted February 28, 2011 Author Share Posted February 28, 2011 Are there any problems changing from ide to ahci. as far as causing problems with data corruption or anything? My boards been in ide for some time now, from memory ahci used to cause problems? Remember you added a thing to disable the ncq at least anyway. Shouldn't be any issue changing to ahci, though if you run pre-5.0-beta5 the "slots" will be wrong. The ncq disable was added because I just observed it had better performance without ncq enabled, but that was quite a few kernels ago and this is something that should be re-visited. Link to comment
Zeron Posted February 28, 2011 Share Posted February 28, 2011 5.0b4 Syslog here: http://pastebin.com/mBCUvBkR Thanks, looking into it. Here is the backtrace if that helps. root@MediaServer1:/boot/debug# gdb /usr/local/sbin/emhttp GNU gdb 6.8 Copyright (C) 2008 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "i486-slackware-linux"... (gdb) run Starting program: /usr/local/sbin/emhttp [Thread debugging using libthread_db enabled] [New Thread 0xb748d6c0 (LWP 1356)] Program received signal SIGSEGV, Segmentation fault. [switching to Thread 0xb748d6c0 (LWP 1356)] 0xb74c396c in ____strtol_l_internal () from /lib/libc.so.6 (gdb) backtrace #0 0xb74c396c in ____strtol_l_internal () from /lib/libc.so.6 #1 0xb74c36d0 in strtol () from /lib/libc.so.6 #2 0x08052901 in disk_spindown_delay () #3 0x08052d14 in check_spindown () #4 0x08056790 in get_status () #5 0x08056a9e in cmdStart () #6 0x08057fc5 in content_init () #7 0x0804e0d1 in init_dyncntnt () #8 0x0804adf8 in main () Link to comment
aiden Posted February 28, 2011 Share Posted February 28, 2011 Shouldn't be any issue changing to ahci, though if you run pre-5.0-beta5 the "slots" will be wrong. I can confirm that changing from IDE to AHCI will renumber the drive "slots" and subsequently put the drives in the wrong order. Link to comment
limetech Posted February 28, 2011 Author Share Posted February 28, 2011 Shouldn't be any issue changing to ahci, though if you run pre-5.0-beta5 the "slots" will be wrong. I can confirm that changing from IDE to AHCI will renumber the drive "slots" and subsequently put the drives in the wrong order. Except no more slots in -beta5 Link to comment
Stokkes Posted February 28, 2011 Share Posted February 28, 2011 I can confirm as well, emhttp segfaults on start up. Suggestions? Feb 27 19:49:53 stewie kernel: emhttp[3495]: segfault at 0 ip b74a096c sp bfa28390 error 4 in libc-2.11.1.so[b7470000+15c000] syslog-20110227.txt Link to comment
bubbaQ Posted February 28, 2011 Share Posted February 28, 2011 Yes, this is pretty nice. Works so far with AHCI and SAS. Did not work with old ICH5 sata controller, but did not look into it much either; may be able to get it to work with non-AHCI. Tom, I have been able to get SATA (and IDE) hot-swapping to work with ICH5, and other non-AHCI mobos, by using scsiadd. http://llg.cubic.org/tools/#scsiadd Link to comment
Whaler_99 Posted February 28, 2011 Share Posted February 28, 2011 Same issue. Replaced the two files on my 5beta4 USB, reboot and no acces. Run emhttp segfaults on start up. Link to comment
limetech Posted February 28, 2011 Author Share Posted February 28, 2011 Ahh hell... you can get -beta5 to start if you go into 'config/disk.cfg', find line that has "spindownDelay" and set as follows: spindownDelay="0" This sets the default spin down delay to "never". I need to take down -beta5 and put up -beta6... Link to comment
dyrewolfe Posted February 28, 2011 Share Posted February 28, 2011 webGui: use hdparm program to query spinning status instead of ioctls()'; now all device type specific code is gone. I guess this hasn't been fully implemented yet? Getting the errors below on the one data disk in the array attached to a BR10i. Querying for temps seem to be working great though. Feb 27 19:01:01 Tower kernel: md: disk2: ATA_OP_SETIDLE1 ioctl error: -22 (Errors) Feb 27 19:08:59 Tower kernel: md: disk2: ATA_OP_SETIDLE1 ioctl error: -22 (Errors) Feb 27 19:10:03 Tower kernel: md: disk2: ATA_OP_SETIDLE1 ioctl error: -22 (Errors) Feb 27 19:47:05 Tower kernel: md: disk2: ATA_OP_STANDBYNOW1 ioctl error: -22 (Errors) Feb 27 20:06:50 Tower kernel: md: disk2: ATA_OP_SETIDLE1 ioctl error: -22 (Errors) Feb 27 20:17:48 Tower kernel: md: disk2: ATA_OP_STANDBYNOW1 ioctl error: -22 (Errors) Feb 27 20:17:50 Tower kernel: md: disk2: ATA_OP_STANDBYNOW1 ioctl error: -22 (Errors) Feb 27 20:17:52 Tower kernel: md: disk2: ATA_OP_STANDBYNOW1 ioctl error: -22 (Errors) Feb 27 20:18:04 Tower kernel: md: disk2: ATA_OP_STANDBYNOW1 ioctl error: -22 (Errors) Feb 27 20:18:14 Tower kernel: md: disk2: ATA_OP_STANDBYNOW1 ioctl error: -22 (Errors) Feb 27 20:18:24 Tower kernel: md: disk2: ATA_OP_STANDBYNOW1 ioctl error: -22 (Errors) Feb 27 20:18:34 Tower kernel: md: disk2: ATA_OP_STANDBYNOW1 ioctl error: -22 (Errors) Feb 27 20:18:44 Tower kernel: md: disk2: ATA_OP_STANDBYNOW1 ioctl error: -22 (Errors) Feb 27 20:18:44 Tower kernel: md: disk2: ATA_OP_STANDBYNOW1 ioctl error: -22 (Errors) Here is output from the console using hdparm to put the drive in standby: root@Tower:~# hdparm -C /dev/sdk /dev/sdk: drive state is: active/idle root@Tower:~# hdparm -y /dev/sdk /dev/sdk: issuing standby command root@Tower:~# hdparm -C /dev/sdk /dev/sdk: drive state is: standby root@Tower:~# Link to comment
queeg Posted February 28, 2011 Share Posted February 28, 2011 It looks like the BR10i should be reporting temps and spinning down now. I'll test it on mine after a little while. Link to comment
Stokkes Posted February 28, 2011 Share Posted February 28, 2011 Ahh hell... you can get -beta5 to start if you go into 'config/disk.cfg', find line that has "spindownDelay" and set as follows: spindownDelay="0" This sets the default spin down delay to "never". I need to take down -beta5 and put up -beta6... Hah, I actually did this before I saw this post seeing as the line before the segfault mentioned the spindowndelay. I figure it was balking at that causing emhttp to choke. Link to comment
limetech Posted February 28, 2011 Author Share Posted February 28, 2011 webGui: use hdparm program to query spinning status instead of ioctls()'; now all device type specific code is gone. I guess this hasn't been fully implemented yet? Getting the errors below on the one data disk in the array attached to a BR10i. Querying for temps seem to be working great though. Right the statement, now all device type specific code is gone is not entirely true, but I wasn't sure if those commands were implemented or not on that controller (I don't have that controller in order to test)... apparently not. This might be a show-stopper as far as that controller is concerned until I re-write the spinup group feature to be user share based, instead of "host group" based. Link to comment
MrD1234 Posted February 28, 2011 Share Posted February 28, 2011 No emhttp segfault here -- I had already set spin down to never. My SAS card controls the spin down. I did get beta5 to hard lock by writing large files from Windows 7 64 over gigabit (22MB/s) and simultaneously hit refresh on the /Main page. I was not able to capture anything and had to kill the VM. Link to comment
Recommended Posts