chickensoup Posted August 8, 2012 Share Posted August 8, 2012 Yes... I think what Tom means is, could you list how to reproduce this error? It will allow other users to try the same thing and see if it is specific to certain hardware/config & help with diagnosing the cause/hopefully a fix. Quote Link to comment
mikejp Posted August 8, 2012 Share Posted August 8, 2012 Tom... Sorry for that last post... the system still had the powerdown-1.02, & a s3 sleep script running I didn't loose the web interface, or have to force a shut down this time... The Call Trace remains... could you list how to reproduce this error? Started a NOCORRECT parity check. Both of my systems exhibit the same behavior. Both are running w/no addons. (now) Configs for both systems are in my sig. Syslogs attached. syslog-LOGO.zip syslog-THOR.zip Quote Link to comment
shlomiassaf Posted August 9, 2012 Share Posted August 9, 2012 After 4 days of hell with REISERFS vs and zam erros I updated to rc6-test2 and it seems to be OK by now (24 hours) I am corssing fingers... Quote Link to comment
BRiT Posted August 18, 2012 Share Posted August 18, 2012 It seems like the patch made it into the SCSI GIT branch with a target slated for Linux 3.6 kernel (hopefully). http://git.kernel.org/?p=linux/kernel/git/jejb/scsi.git;a=commit;h=98dc81b0d6c483a3eb256764ae10f156ccefdbbb Quote Link to comment
savestheday Posted August 20, 2012 Share Posted August 20, 2012 Question....I just upgraded all of my LSI cards to P14 in order to upgrade to this test RC. If I upgrade and for some reason need to go back, what will happen to the extra drives I'm able to add with this test version? I'm guessing parity will have to be rebuilt but what are the other ramifications? Quote Link to comment
pras1011 Posted August 24, 2012 Share Posted August 24, 2012 Why is the parity checks and write speeds slower on this than RC5? RC5 parity check speed is 75MB/s and Test2 is 45MB/s. Quote Link to comment
BRiT Posted August 24, 2012 Share Posted August 24, 2012 The Linux Kernel version used, possibly having to do with IO priority and control groups for processes. Quote Link to comment
Frank1940 Posted August 24, 2012 Share Posted August 24, 2012 Why is the parity checks and write speeds slower on this than RC5? RC5 parity check speed is 75MB/s and Test2 is 45MB/s. Did the actual time to complete a parity check change? Quote Link to comment
pras1011 Posted August 25, 2012 Share Posted August 25, 2012 Yes the RC5 was 14 hours and Test2 was 26 hours! Quote Link to comment
moose Posted August 27, 2012 Share Posted August 27, 2012 I setup a test server running v5.0-rc6-r8168-test2. Only plugin is unmenu. I have a 3TB parity and 2TB data disk. The 3TB disk does not spin down (constant green activity LED). The 2 disks are in a 5-in-3 cage and I've moved the 3TB to several different slots and had the same result...no spin down. Both drives are using interfacing via a Super Micro AOC-SAS2LP-MV8 controller. 2TB spins down fine. Any suggestions? The specs for this rig are in my signature...it's the test server. syslog and smart report for subject server/drive are attached. syslog-2012-08-26.txt Smart_report.txt Quote Link to comment
Frank1940 Posted August 27, 2012 Share Posted August 27, 2012 I setup a test server running v5.0-rc6-r8168-test2. Only plugin is unmenu. I have a 3TB parity and 2TB data disk. The 3TB disk does not spin down (constant green activity LED). The 2 disks are in a 5-in-3 cage and I've moved the 3TB to several different slots and had the same result...no spin down. Both drives are using interfacing via a Super Micro AOC-SAS2LP-MV8 controller. 2TB spins down fine. Any suggestions? The specs for this rig are in my signature...it's the test server. syslog and smart report for subject server/drive are attached. Be careful as there are several reports that the lights on drive cages are not accurate or right, depending on your viewpoint! I have two Seagate 3TB drives in Monoprice 'hot swap' bays and the lights are always on-- yet the drives shut down! Quote Link to comment
moose Posted August 27, 2012 Share Posted August 27, 2012 Be careful as there are several reports that the lights on drive cages are not accurate or right, depending on your viewpoint! I have two Seagate 3TB drives in Monoprice 'hot swap' bays and the lights are always on-- yet the drives shut down! ok, if this is the case, do I trust the unRAID web-interface which has a blinking green status (spun-down) for the 3TB drive versus the drive cage LED (solid green)? I'll search for reports of inaccurate drive cage disk status LEDs, specifically for the Super Micro CSE-M35T-1B. Quote Link to comment
JonathanM Posted August 27, 2012 Share Posted August 27, 2012 ok, if this is the case, do I trust the unRAID web-interface which has a blinking green status (spun-down) for the 3TB drive versus the drive cage LED (solid green)? If you have halfway decent hearing, I'd trust my ears. Try an old mechanics trick, by holding a solid bar to the face of the tray and the bones near your ear, you should be able to hear the motor spin up the drive quite clearly. Listen to the cage as you access a supposedly spun down drive and see if you hear it spin up. Noisy fans may make it a little tougher to hear, but you should be able to hear something. Also, the temperature of a spun down drive should be lower than an active drive. Quote Link to comment
Frank1940 Posted August 27, 2012 Share Posted August 27, 2012 Be careful as there are several reports that the lights on drive cages are not accurate or right, depending on your viewpoint! I have two Seagate 3TB drives in Monoprice 'hot swap' bays and the lights are always on-- yet the drives shut down! ok, if this is the case, do I trust the unRAID web-interface which has a blinking green status (spun-down) for the 3TB drive versus the drive cage LED (solid green)? I'll search for reports of inaccurate drive cage disk status LEDs, specifically for the Super Micro CSE-M35T-1B. Yes, I would trust the interface more than the drive cage lights. You can use the metal rod technique of the previous poster. You could also access the spundown drive to play a file as it takes a few seconds for a spundown drive to spin up. Then do the same for a different file after this first test while the drive is still spun up. By the way, you can spin the drives up and down using the Simple Features interface and it would not surprise me that the you can do so on the standard interface. (It is the up or down arrows under the "Device" heading on the Main page.) Quote Link to comment
drew.c Posted August 28, 2012 Share Posted August 28, 2012 Be careful as there are several reports that the lights on drive cages are not accurate or right, depending on your viewpoint! I have two Seagate 3TB drives in Monoprice 'hot swap' bays and the lights are always on-- yet the drives shut down! ok, if this is the case, do I trust the unRAID web-interface which has a blinking green status (spun-down) for the 3TB drive versus the drive cage LED (solid green)? I'll search for reports of inaccurate drive cage disk status LEDs, specifically for the Super Micro CSE-M35T-1B. I had a solid red light problem with my Seagate 3TB and my "Icy Dock 5 in 3". Upgrading the firmware of the drive fixed the problem. http://lime-technology.com/forum/index.php?topic=21061.msg187057#msg187057 Drew Quote Link to comment
moose Posted August 28, 2012 Share Posted August 28, 2012 Thank you drew.c! This was the problem. I found a similar post (prior to your reply) http://lime-technology.com/forum/index.php?topic=22201.0 Upgrading the Seagate ST3000DM001-9YN166 parity drive firmware from CC46 to CC4H solved the LED problem on the CSE-M35T cage! I believe the LED was operating in "inverse" mode, meaning that the LED was solid when the drive was inactive and off when the drive was active, upgrading the drive firmware reversed the LED behavior to normal. (Thanks also to Frank1940 and jonathanm for idea of using the metal bar to "listen", however there are so many fans running in the room where the test server is located, I couldn't hear the subtle noise of the motor....great idea however.) Since this topic did not specifically apply to 5.0-rc6-r8168-test2, but was resolved with a drive firmware update, the moderators might want to move it to another area. Quote Link to comment
Thornwood Posted September 4, 2012 Share Posted September 4, 2012 Sorry to bother, But are we waiting for another Kernel again? We were on a role there for a while and now it feels like we are on the back burner.... I could be wrong just how I feel. Thank you Quote Link to comment
Frank1940 Posted September 4, 2012 Share Posted September 4, 2012 Sorry to bother, But are we waiting for another Kernel again? We were on a role there for a while and now it feels like we are on the back burner.... I could be wrong just how I feel. Thank you I believe that is the case. In one of the threads in the test2 release was being discussed, the 'fix' that worked to repair the last major obstacle in the release of version 5 was to be included in the 3.6 release of the kernel. Of course, this was not actually posted by Tom but was inferred from some of his comments and another user's comment on the status of the Linux kernel releases. Quote Link to comment
sam_roberts30 Posted September 4, 2012 Share Posted September 4, 2012 Is there somewhere to get a list of outstanding bugs that need to be over come before 5 final? Quote Link to comment
Frank1940 Posted September 5, 2012 Share Posted September 5, 2012 Is there somewhere to get a list of outstanding bugs that need to be over come before 5 final? As you should have realized by now, such a list does not exist anywhere on the Internet. As far as I know from my reading of the posts on this site, there are no major issues left in the latest release candidate (V5.0-rc6-r8168-test2). The rc was released on July 26th. You can read the posts after that date if you want to see if there might be some minor thing that would cause you not to want to update. But the only things that I have have seen have been cockpit errors... Which brings up a second point. (And this is not directed only at you!) If you have a working mature unRAID system running on ver4.X, why are considering upgrading? You should be asking yourself that before you even consider the upgrade path. If there is some feature that you need, great, go to it. If it is just because it is there and your system is critical to your lifestyle (or business) or you don't have the time to deal with minor hassles, maybe you should not upgrade at all! There is an old adage, the version X.0 of any software is merely the last beta and the last group of beta testers are the early adapters! Quote Link to comment
dheg Posted September 6, 2012 Share Posted September 6, 2012 Is there somewhere to get a list of outstanding bugs that need to be over come before 5 final? As you should have realized by now, such a list does not exist anywhere on the Internet. As far as I know from my reading of the posts on this site, there are no major issues left in the latest release candidate (V5.0-rc6-r8168-test2). The rc was released on July 26th. You can read the posts after that date if you want to see if there might be some minor thing that would cause you not to want to update. But the only things that I have have seen have been cockpit errors... Which brings up a second point. (And this is not directed only at you!) If you have a working mature unRAID system running on ver4.X, why are considering upgrading? You should be asking yourself that before you even consider the upgrade path. If there is some feature that you need, great, go to it. If it is just because it is there and your system is critical to your lifestyle (or business) or you don't have the time to deal with minor hassles, maybe you should not upgrade at all! There is an old adage, the version X.0 of any software is merely the last beta and the last group of beta testers are the early adapters! +1 Sent from my GT-I9100 using Tapatalk 2 Quote Link to comment
BRiT Posted September 15, 2012 Share Posted September 15, 2012 Linux Kernel 3.5.4 includes the MPT2SAS FIX! http://www.kernel.org/pub/linux/kernel/v3.0/ChangeLog-3.5.4 It was just released today, September 14th. Quote Link to comment
pras1011 Posted September 15, 2012 Share Posted September 15, 2012 Linux Kernel 3.5.4 includes the MPT2SAS FIX! http://www.kernel.org/pub/linux/kernel/v3.0/ChangeLog-3.5.4 It was just released today, September 14th. Can someone install this into one of the RCs? Quote Link to comment
JackBauer Posted September 15, 2012 Share Posted September 15, 2012 Yes... Tom can and will, I'd almost bet my life on it... And he'll probably do it quickly. Do some internal testing, then release it for the community to test. I do believe he's doing everything he can, with the exception of communicating what is going on. 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.