Re: LSI supported mps(4) driver available
On Apr 4, 2012 10:02 PM, Matt Thyer matt.th...@gmail.com wrote: On 3 April 2012 23:12, Gary Palmer gpal...@freebsd.org wrote: I think you should contact either SuperMicro or LSI and open a support case as it looks like there could be a problem with either the controller or the firmware when presented with mixed speed devices. Either way I think this needs to be escalated to the manufacturer. Regards, Gary I'm now having no problems since moving the SATA 3 drive to the on board Intel controller. I'll try to report this to Super Micro LSI. I spoke too soon. The problem of the SATA 3 drive being FAULTED in the raidz2 pool has indeed been solved by moving that drive from the Super micro (SAS 6G) controller to the onboard Intel (SATA 2) controller. However, the 157k interrupts per second problem remained (its not apparent immediately after boot). However, even this problem has been resolved by upgrading from 8-STABLE to 9-STABLE (as I reported in the freebsd-stable list). So I'm happy now but still no closer to understanding the cause. I'm guessing that it was either USB related or something to do with the on CPU package Intel graphics of the Core i3 530 CPU. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: LSI supported mps(4) driver available
On 3 April 2012 23:12, Gary Palmer gpal...@freebsd.org wrote: On Tue, Apr 03, 2012 at 10:52:25PM +0930, Matt Thyer wrote: I forgot to mention that I'm still having problems after this phase 11 firmware upgrade with the 6 Gbps drive being kicked out of the raidz2 with write errors (even though a SMART full surface test says the drive is OK). This leads me to think that both the old and new drivers have a problem with the 6 Gbps WD20EARX-00P AB51 drive. Now that the 6 Gbps drive is on the Intel SATA controller things seem OK but it's a bit early to tell. Stay tuned! I think you should contact either SuperMicro or LSI and open a support case as it looks like there could be a problem with either the controller or the firmware when presented with mixed speed devices. Either way I think this needs to be escalated to the manufacturer. Regards, Gary I'm now having no problems since moving the SATA 3 drive to the on board Intel controller. I'll try to report this to Super Micro LSI. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: LSI supported mps(4) driver available
On 28 March 2012 03:51, Kenneth D. Merry k...@freebsd.org wrote: On Tue, Mar 27, 2012 at 23:50:31 +1030, Matt Thyer wrote: On 26 March 2012 23:55, Gary Palmer gpal...@freebsd.org wrote: On Mon, Mar 26, 2012 at 08:05:59PM +1030, Matt Thyer wrote: On Mar 26, 2012 3:43 AM, Garrett Cooper yaneg...@gmail.com wrote: On Sun, Mar 25, 2012 at 5:16 AM, Matt Thyer matt.th...@gmail.com wrote: Has this driver been MFC to 8-STABLE yet ? I'm asking because I updated my NAS on the 4th of March from 8-STABLE r225723 to r232477 and am now seeing 157,000 interrupts per second on irq 16 where my SuperMicro AOC-USAS2-L8i resides (this card uses the LSI SAS2008 chip). [snip] After encountering this problem I updated my firmware from phase 7 to phase 11 but this did not fix things. My question is: Is the LSI driver even in 8-STABLE yet?. If not I'll upgrade to 9-STABLE to get the new driver. If it is, then I want to downgrade to just before it came in to see if this high interrupt rate problem is fixed. I'm no export in svn, however: http://svnweb.freebsd.org/base?view=revisionamp;revision=230922 would appear to suggest that the new driver is in 8-Stable Gary It's painful to take this system back to r230921 due to intolerance for downtime from it's users so I'd like to investigate the cause of the problem and try patches/sysctls/whatever first. The drives I'm using are 7 x WDC WD20EARS-00M (3 are AB50, 4 are AB51) and 1 x WD20EARX-00P AB51. The WD20EARX-00P AB51 is a SATA 3 (6 Gbps) drive but the others are all SATA 2 (3 Gbps). I know the driver doesn't like mixed speeds in IR mode but I'm flashed with IT firmware as ZFS is doing my RAID (raidz2). I was having problems with the WD20EARX-00P AB51 drive being faulted by ZFS until I updated the firmware to 11 and now ZFS is happy (I've also done a full extended drive SMART test and the drive is fine). So what do people suggest (before reversion to r230921) ? If you're going to prove that it's the new LSI driver, you will probably have to go back to the old driver. You don't have to back out your entire tree, you can just back out the driver itself if you have an SVN tree. You can go into sys/dev/mps and do: svn update -r 230714 And then edit sys/conf/files and comment out these three lines: dev/mps/mps_config.coptional mps dev/mps/mps_mapping.c optional mps dev/mps/mps_sas_lsi.c optional mps Then you should be able to rebuild your kernel with the old driver and see if the problem occurs again. Ken -- Kenneth Merry k...@freebsd.org This didn't work for me so I removed my /usr/src and checked out 8-STABLE at revision 230921 (svn checkout -r 230921 http://svn.freebsd.org/base/stable/8 /usr/src). I've built world, kernel etc and installed it using GENERIC kernel done my mergemaster, delete old, delete old-libs and I still have the problem. I'm wondering if it's due to the single 6 Gb drive in my raidz2 (the other 7 are 3 Gb). I've heard that the new driver doesn't like mixed speeds in a raid set when using -IR firmware but I wouldn't expect an issue with ZFS with -IT firmware. It seems that there may be a general incompatibility with both the old and new drivers and the Western Digital WD20EARX-00P 6 Gbps drive. Unfortunately I cannot get the old 3 Gb drive anymore. I'll try moving the WD20EARX-00P drive to the on board SATA ports next. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: LSI supported mps(4) driver available
On Mar 27, 2012 11:50 PM, Matt Thyer matt.th...@gmail.com wrote: I was having problems with the WD20EARX-00P AB51 drive being faulted by ZFS until I updated the firmware to 11 and now ZFS is happy (I've also done a full extended drive SMART test and the drive is fine). I forgot to mention that I'm still having problems after this phase 11 firmware upgrade with the 6 Gbps drive being kicked out of the raidz2 with write errors (even though a SMART full surface test says the drive is OK). This leads me to think that both the old and new drivers have a problem with the 6 Gbps WD20EARX-00P AB51 drive. Now that the 6 Gbps drive is on the Intel SATA controller things seem OK but it's a bit early to tell. Stay tuned! ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Using TMPFS for /tmp and /var/run?
On Mar 30, 2012 6:22 AM, Eric van Gyzen e...@vangyzen.net wrote: However, if you always want to use tmpfs instead of stable storage, please do not. Some people expect /tmp to be persistent. This is why /etc/defaults/rc.conf has clear_tmp_enable=NO. Changing this would break the POLA. This is a mistake. The default should be clear_tmp_enable=YES if only to uncover those broken configurations that expect /tmp to be persistent. I was going to say At least on -CURRENT but in reality it should be the default on -RELEASE too when it's clearly a bug to expect /tmp to be persistent. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: LSI supported mps(4) driver available
On 26 March 2012 23:55, Gary Palmer gpal...@freebsd.org wrote: On Mon, Mar 26, 2012 at 08:05:59PM +1030, Matt Thyer wrote: On Mar 26, 2012 3:43 AM, Garrett Cooper yaneg...@gmail.com wrote: On Sun, Mar 25, 2012 at 5:16 AM, Matt Thyer matt.th...@gmail.com wrote: Has this driver been MFC to 8-STABLE yet ? I'm asking because I updated my NAS on the 4th of March from 8-STABLE r225723 to r232477 and am now seeing 157,000 interrupts per second on irq 16 where my SuperMicro AOC-USAS2-L8i resides (this card uses the LSI SAS2008 chip). [snip] After encountering this problem I updated my firmware from phase 7 to phase 11 but this did not fix things. My question is: Is the LSI driver even in 8-STABLE yet?. If not I'll upgrade to 9-STABLE to get the new driver. If it is, then I want to downgrade to just before it came in to see if this high interrupt rate problem is fixed. I'm no export in svn, however: http://svnweb.freebsd.org/base?view=revisionamp;revision=230922 would appear to suggest that the new driver is in 8-Stable Gary It's painful to take this system back to r230921 due to intolerance for downtime from it's users so I'd like to investigate the cause of the problem and try patches/sysctls/whatever first. The drives I'm using are 7 x WDC WD20EARS-00M (3 are AB50, 4 are AB51) and 1 x WD20EARX-00P AB51. The WD20EARX-00P AB51 is a SATA 3 (6 Gbps) drive but the others are all SATA 2 (3 Gbps). I know the driver doesn't like mixed speeds in IR mode but I'm flashed with IT firmware as ZFS is doing my RAID (raidz2). I was having problems with the WD20EARX-00P AB51 drive being faulted by ZFS until I updated the firmware to 11 and now ZFS is happy (I've also done a full extended drive SMART test and the drive is fine). So what do people suggest (before reversion to r230921) ? ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: LSI supported mps(4) driver available
On Mar 26, 2012 3:43 AM, Garrett Cooper yaneg...@gmail.com wrote: On Sun, Mar 25, 2012 at 5:16 AM, Matt Thyer matt.th...@gmail.com wrote: On 21 January 2012 09:58, Kenneth D. Merry k...@freebsd.org wrote: On Fri, Jan 20, 2012 at 23:14:20 -, Steven Hartland wrote: - Original Message - From: Kenneth D. Merry k...@freebsd.org To: freebsd-s...@freebsd.org; freebsd-current@freebsd.org Sent: Friday, January 20, 2012 8:44 PM Subject: LSI supported mps(4) driver available The LSI-supported version of the mps(4) driver that supports their 6Gb SAS HBAs as well as WarpDrive controllers, is available here: http://people.freebsd.org/~ken/lsi/mps_lsi.20120120.1.txt I plan to check it in to head next week, and then MFC it into stable/9 a week after that most likely. Great to see this being done, thanks to everyone! Be even better to see this MFC'ed to 8.x as well if all goes well. Do you think this will possible? Yes, that should be doable as well. It's unlikely that all of the CAM changes will get merged back, but the driver itself shouldn't be a problem. Ken Has this driver been MFC to 8-STABLE yet ? I'm asking because I updated my NAS on the 4th of March from 8-STABLE r225723 to r232477 and am now seeing 157,000 interrupts per second on irq 16 where my SuperMicro AOC-USAS2-L8i resides (this card uses the LSI SAS2008 chip). More details are in a thread on the freebsd-stable mailing list entitled 157k interrupts per second causing 60% CPU load on idle system. The first message is here: http://www.freebsd.org/cgi/getmsg.cgi?fetch=152290+156717+/usr/local/www/db/text/2012/freebsd-stable/20120325.freebsd-stable If this new driver isn't in 8-STABLE yet I think I'll try upgrading the whole system to 9-STABLE. Be sure to update your firmware beforehand. v11 firmware from LSI (or the OEM vendor) is required in order for all drives to be detected in FreeBSD in certain configs. Cheers, -Garrett After encountering this problem I updated my firmware from phase 7 to phase 11 but this did not fix things. My question is: Is the LSI driver even in 8-STABLE yet?. If not I'll upgrade to 9-STABLE to get the new driver. If it is, then I want to downgrade to just before it came in to see if this high interrupt rate problem is fixed. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: LSI supported mps(4) driver available
On 21 January 2012 09:58, Kenneth D. Merry k...@freebsd.org wrote: On Fri, Jan 20, 2012 at 23:14:20 -, Steven Hartland wrote: - Original Message - From: Kenneth D. Merry k...@freebsd.org To: freebsd-s...@freebsd.org; freebsd-current@freebsd.org Sent: Friday, January 20, 2012 8:44 PM Subject: LSI supported mps(4) driver available The LSI-supported version of the mps(4) driver that supports their 6Gb SAS HBAs as well as WarpDrive controllers, is available here: http://people.freebsd.org/~ken/lsi/mps_lsi.20120120.1.txt I plan to check it in to head next week, and then MFC it into stable/9 a week after that most likely. Great to see this being done, thanks to everyone! Be even better to see this MFC'ed to 8.x as well if all goes well. Do you think this will possible? Yes, that should be doable as well. It's unlikely that all of the CAM changes will get merged back, but the driver itself shouldn't be a problem. Ken Has this driver been MFC to 8-STABLE yet ? I'm asking because I updated my NAS on the 4th of March from 8-STABLE r225723 to r232477 and am now seeing 157,000 interrupts per second on irq 16 where my SuperMicro AOC-USAS2-L8i resides (this card uses the LSI SAS2008 chip). More details are in a thread on the freebsd-stable mailing list entitled 157k interrupts per second causing 60% CPU load on idle system. The first message is here: http://www.freebsd.org/cgi/getmsg.cgi?fetch=152290+156717+/usr/local/www/db/text/2012/freebsd-stable/20120325.freebsd-stable If this new driver isn't in 8-STABLE yet I think I'll try upgrading the whole system to 9-STABLE. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: growfs remove ufs/label and can't reset it with tunefs
2012/3/9 Olivier Cochard-Labbé oliv...@cochard.me Hi all, once run growfs on a partition that had an UFS label, this label is removed and it's no more possible to re-set it with tunefs. Here is how to reproduce (tested on 8.3 and 9.0): mdconfig -a -t malloc -s 10MB gpart create -s mbr /dev/md0 gpart add -t freebsd -s 5MB /dev/md0 newfs -L THELABEL /dev/md0s1 glabel status | grep THELABEL = Label is present, now we resize the slice: gpart resize -i 1 /dev/md0 glabel status | grep THELABEL = Label is still present, now we growfs the slice: growfs /dev/md0s1 glabel status | grep THELABEL = UFS label disapear ! Ok, I will try to re-set it: tunefs -L THELABEL /dev/md0s1 glabel status | grep THELABEL = Still no label !?! Should I create a PR about this problem ? Regards, Olivier Yes, It is important to record this problem in the PR system. I suspect that the problem is with growfs as it needs to be taught to not overwrite the end of the volume where the label information is stored. (It will need to examine the volume to see if GEOM has information stored at the end of the volume such that the grow should not overwrite the GEOM metadata). Matthew ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Enhancing the user experience with tcsh
How about adding stuff like this to /usr/share/examples/tcsh/complete.tcsh ? -- Joel Yes to that. This is exactly where these suggestions should go. Feel free to create multiple examples files there but be very carefully with changes to system wide defaults. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: [HEADSUP][CFT] pkgng beta1 is out
On Feb 6, 2012 3:50 AM, Radio młodych bandytów radiomlodychbandy...@o2.pl wrote: I wonder if I'm the only one thinking about a decentralised package management First, a decentralised transport layer. Torrents are faster and more reliable than servers. Second, decentralised management when anybody can upload a port directly into the system. -- Twoje radio Such a system would need to support traditional protocols such as FTP HTTP due to many corporate environments not allowing anything else. I'm all for a distributed system but you can't forget the corporate users. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Update on ports on 10.0
On Oct 11, 2011 5:07 PM, Erwin Lansing er...@freebsd.org wrote: Since the release has been pushed back some more since the last mail, we do have some time to test a possible fix for the issues we're seeing with libtool on FreeBSD 10.0. [snip] to move forward. Other options include the big find/grep/awk solution that has been posted several times and fiddling with uname to go to FreeBSD 9.99 for a while, while ports can be fixed. I would vote for a prominent message in the places people who run -CURRENT are meant to pay attention to (this mailing list and UPDATING and the ports equivalents). The message should point people to a web site (wiki?) with the latest news on how to report and provide patches for fixes with guidance on typical solutions. After all, this is -CURRENT and it's users are meant to be contributing fixes and are not expected to require their hands to be held. This I believe would be a good method to leverage the community so as to distribute the fixup problem. In particular the hack of 9.99 is likely to cause more problems and should be avoided. My 2¢ ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: System headers with clang?
On Oct 12, 2011 3:25 AM, Larry Rosenman l...@lerctr.org wrote: I didn't say bug for bug, just not generate stupid errors like the ffs one. -- Sent from my Android phone with K-9 Mail. Please excuse my brevity. Chuck Swiger cswi...@mac.com wrote: On Oct 11, 2011, at 6:59 AM, Larry Rosenman wrote: We will NOT support clang as the compiler for lsof unless the system headers work the same way as gcc's do. That apparently means you won't support clang then, because it's not intended to be (or ever going to be) fully bug-for-bug compatible with GCC. In this case, at least, clang is reporting legitimate issues which should be fixed, even if folks continue to build lsof with GCC from now until the end of days. The elegant solution would be to avoid this problem altogether by re-implementation of lsof using interfaces into the kernel that provide the required information. bsdof anyone? ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Memstick image differences between 8.x and 9.x
On 9 October 2011 19:54, Thomas Mueller mueller6...@bellsouth.net wrote: One issue that has not come up on the emailing list is that dd, designed to work with memsticks of various capacities, can not make the backup gpt at the end of the memstick. Partition is just big enough to hold the data, and I ran out of inodes at times due to the installer writing to /tmp on the memstick. Maybe a script to put a backup GPT at the end on the memstick, and make the second partition fill the space available so as not to be too tightly squeezed? A script is a kludge. As we have no need for larger than 2TiB install images we should use the POLA and continue to support dd writing our memstick images which means don't use GPT for them. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Memstick image differences between 8.x and 9.x
On 9 October 2011 13:20, Thomas K. f...@gothschlampen.com wrote:
On Sat, Oct 08, 2011 at 07:28:55PM -0600, Warren Block wrote:
On Sun, 9 Oct 2011, Matt Thyer wrote:
On Oct 9, 2011 11:04 AM, Nathan Whitehorn nwhiteh...@freebsd.org
wrote:
On 10/08/11 19:25, Matt Thyer wrote:
There is also the interesting question of actually installing to GPT on
the hard disk, which is the default in 9.0. Does this not work on some
systems? If so, do we want to blacklist
them and use a different default partition scheme? Can we identify
systems that violate regular PC boot standards and reject GPT? Any data on
any of these points would be appreciated.
I don't think there have been any reports of failure to boot properly
formatted GPT yet.
Lenovo T420S and T520, from the links above. Install GPT on the
hard drive, try to boot.
http://forums.freebsd.org/showthread.php?t=26304
http://forums.freebsd.org/showthread.php?t=26759
http://forum.thinkpads.com/viewtopic.php?f=9t=98078
As I used parted from Linux to fix the alternate GPT, i.e. put it not at
the
end of the image data but on the end of the disk, and it still did not
appear
in the boot device list, the Acer AX3960 should probably be on the list
as well.
Being a Core i7 2600k system maybe 6 months old, it's rather recent
hardware,
but doesn't boot from the memstick image.
Regards,
Thomas
Failure to boot the FreeBSD 9.0-BETA{2|3} memstick images does not indicate
a problem with a PCs BIOS/UEFI as these images are not properly formatted.
If we were able to come up with examples of BIOS/UEFI that cannot boot from
GPT partitioned volumes there would not be a problem as long as bsdinstall
still supports partitioning volumes with MSDOS/MBR partitioning schemes.
The big problem is being able to launch the installation process to start
with which is yet another reason to have the memstick image non-GPT even if
you could work out a script/kludge etc to be able to write a properly
formatted GPT memstick.
The solution to this issue is obvious.
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Memstick image differences between 8.x and 9.x
On 9 October 2011 21:44, Adrian Chadd adr...@freebsd.org wrote:
On 9 October 2011 19:10, Matt Thyer matt.th...@gmail.com wrote:
Failure to boot the FreeBSD 9.0-BETA{2|3} memstick images does not
indicate
a problem with a PCs BIOS/UEFI as these images are not properly
formatted.
Accepted.
If we were able to come up with examples of BIOS/UEFI that cannot boot
from
GPT partitioned volumes there would not be a problem as long as
bsdinstall
still supports partitioning volumes with MSDOS/MBR partitioning schemes.
The big problem is being able to launch the installation process to start
with which is yet another reason to have the memstick image non-GPT even
if
you could work out a script/kludge etc to be able to write a properly
formatted GPT memstick.
The solution to this issue is obvious.
Yes, it's the current solution has a lot of unknown-how broken stuff
about it, let's revert it for 9.0 and then use the 10.0 release cycle
to do further research and testing.
Unfortunately there is no reasonable revert path here. bsdinstall is the
way forward and I agree it should be the installer for 9.0-RELEASE.
Currently bsdinstall relies on labels and that's a good thing (intelligent
design choice).
Work is already underway to make the memstick issue with UFS labels and
MSDOS/MBR partitioning and when that's done this issue will be solved.
So it's not a matter of reverting, it's a matter of forging ahead and
delaying the release as this is a show stopper.
___
freebsd-current@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Memstick image differences between 8.x and 9.x
On Oct 9, 2011 8:52 AM, Warren Block wbl...@wonkity.com wrote: On Sat, 8 Oct 2011, Glen Barber wrote: On 10/8/11 5:40 PM, Warren Block wrote: On Sat, 8 Oct 2011, Glen Barber wrote: On 10/8/11 2:21 PM, Garrett Cooper wrote: Are there any general structural differences between FreeBSD 8 and 9 memstick images which could be at fault here? The new memstick image uses GPT instead of MBR partitioning. GPT should have no impact on booting from the memory stick, as far as I am aware. Memory stick should not be a problem, but some of the Lenovo notebooks hate GPT, even with a PMBR: http://forums.freebsd.org/showthread.php?t=26304 http://forums.freebsd.org/showthread.php?t=26759 Ugh, that's annoying. I'm half-tempted to note this in the new installer chapter, but I don't like the idea of such edge cases as these to effectively turn that page into a pseudo-HCL. There are already a couple of notes about having to use MBR with XP and other older operating systems. But instead of updating them, I'd rather see somebody with one of the affected systems contact somebody with influence at Lenovo and say hey, the FreeBSD guys are talking about making your broken GPT support famous followed quickly by a BIOS update. I believe this is actually a case of the memstick image being an improperly formatted GPT as there is no backup partition table at the end of the volume. The only sensible answer is to not use GPT for the memstick image. I not said this,loud enough yet but this is a show stopper for 9.0-RELEASE and must be fixed. We can't have a major release that modern systems cannot install with one of now most popular install methods. As a first step, Andriy Gapon has provided a quick patch for makefs(8) so it can create filesystems with UFS labels (as bsdinstall relys on labels). If you want to fix your memstick, create a copy of the partition table at the end of the volume and it should boot. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Memstick image differences between 8.x and 9.x
On Oct 9, 2011 11:04 AM, Nathan Whitehorn nwhiteh...@freebsd.org wrote: On 10/08/11 19:25, Matt Thyer wrote: I believe this is actually a case of the memstick image being an improperly formatted GPT as there is no backup partition table at the end of the volume. The only sensible answer is to not use GPT for the memstick image. I've not said this loud enough yet but this is a show stopper for 9.0-RELEASE and must be fixed. We can't have a major release that modern systems cannot install with one of the now most popular install methods. As a first step, Andriy Gapon has provided a quick patch for makefs(8) so it can create filesystems with UFS labels (as bsdinstall relys on labels). If you want to fix your memstick, create a copy of the partition table at the end of the volume and it should boot. It is being fixed, pending Andriy's change getting into the tree, which should be soon, and will end up being used for the next build (which I believe is RC1). There is also the interesting question of actually installing to GPT on the hard disk, which is the default in 9.0. Does this not work on some systems? If so, do we want to blacklist them and use a different default partition scheme? Can we identify systems that violate regular PC boot standards and reject GPT? Any data on any of these points would be appreciated. -Nathan I don't think there have been any reports of failure to boot properly formatted GPT yet. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Strange ZFS filesystem corruption
On Oct 4, 2011 5:50 AM, Paul Mather p...@gromit.dlib.vt.edu wrote: [snip] I have a raidz2 ZFS pool on a system that I have recently been using as a mirror for about 6.5 TiB of data. The data are mirrored nightly using rsync. I noticed during these nightly rsync copies I would get some errors like this: = file has vanished: /backups/storage/san/DLA/DLA_Records/05DLAAdmin rsync: stat /backups/storage/san/DLA/DLA_Records/05DLAAdmin failed: No such file or directory (2) rsync: recv_generator: mkdir /backups/storage/san/DLA/DLA_Records/05DLAAdmin/05DI_business copy failed: No such file or directory (2) *** Skipping any contents from this failed directory *** [snip] I know ZFS does not have a fsck utility (because it doesn't need one:), but does anyone know of any way of fixing this corruption short of destroying the pool, creating a new one, and restoring from backup? Is there some way of exporting and re-importing the pool that has the side-effect of doing some kind of fsck-like repairing of subtle corruption like this? Cheers, Paul. Paul, I'd be looking at the hardware or drivers. What disk model are you using and what controller? Are you using port multipliers and are there known issues with your hardware combination? ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: FreeBSD 9.0-BETA2 memstick USB image hangs my BIOS
On Sep 26, 2011 2:33 PM, Adrian Chadd adr...@freebsd.org wrote: On 26 September 2011 12:56, Kevin Oberman kob6...@gmail.com wrote: I suspect that the thumb drive is the issue, not FreeBSD. I've booted the drive successfully on an older Via C3 board. I'm now watching it boot successfully on a Gigabyte GA-8I915PC Duo board. I'm going to try dumping a Linux distro on this USB disk after I get PCBSD 9.0-BETA2 installed. That way I can try to eliminate the USB disk as being funny. (Although it's quite possible the USB disk is doing other funny things.) I really do think that this BIOS of mine dislikes a GPT partition inside an MBR/DOS setup. If this is the case, I wonder how many other motherboard/BIOSes have this problem. Adrian I believe this would be due to the improper GPT partition table on the FreeBSD memstick (as there is no backup partition table at the end of the volume). This was discussed earlier and now that makefs has patches for UFS label support there should be no need to continue to use GPT partitioning for the memstick image. The background is that bsdinstall relys on labels (a good thing) and GPT partition labels were being used but as there's no way to predict the size of your memstick, the resulting GPT partition table was not valid (for strict UEFI systems). The fix was to go back to traditional MS-DOS partitioning (MBR) but this couldn't be done until makefs supported UFS labels. Someone came up with patches so 9.0 should be able to have an MBR memstick and bsdinstall. I'm not sure how far this has progressed. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: FreeBSD 9.0-BETA2 memstick USB image hangs my BIOS
On Sep 28, 2011 4:11 AM, Nathan Whitehorn nwhiteh...@freebsd.org wrote: On 09/27/11 08:24, Matt Thyer wrote: On Sep 26, 2011 2:33 PM, Adrian Chaddadr...@freebsd.org wrote: On 26 September 2011 12:56, Kevin Obermankob6...@gmail.com wrote: I suspect that the thumb drive is the issue, not FreeBSD. I've booted the drive successfully on an older Via C3 board. I'm now watching it boot successfully on a Gigabyte GA-8I915PC Duo board. I'm going to try dumping a Linux distro on this USB disk after I get PCBSD 9.0-BETA2 installed. That way I can try to eliminate the USB disk as being funny. (Although it's quite possible the USB disk is doing other funny things.) I really do think that this BIOS of mine dislikes a GPT partition inside an MBR/DOS setup. If this is the case, I wonder how many other motherboard/BIOSes have this problem. Adrian I believe this would be due to the improper GPT partition table on the FreeBSD memstick (as there is no backup partition table at the end of the volume). This was discussed earlier and now that makefs has patches for UFS label support there should be no need to continue to use GPT partitioning for the memstick image. The background is that bsdinstall relys on labels (a good thing) and GPT partition labels were being used but as there's no way to predict the size of your memstick, the resulting GPT partition table was not valid (for strict UEFI systems). The fix was to go back to traditional MS-DOS partitioning (MBR) but this couldn't be done until makefs supported UFS labels. Someone came up with patches so 9.0 should be able to have an MBR memstick and bsdinstall. I'm not sure how far this has progressed. I don't believe these patches were ever committed. Could you provide a pointer to them? -Nathan Yes, it was Andriy Gapon in this message: http://www.FreeBSD.org/cgi/mid.cgi?db=irtid=4e60f480.6040...@freebsd.org ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: FreeBSD 9.0-BETA2 memstick USB image hangs my BIOS
On Sep 28, 2011 7:15 AM, Matt Thyer matt.th...@gmail.com wrote: On Sep 28, 2011 4:11 AM, Nathan Whitehorn nwhiteh...@freebsd.org wrote: On 09/27/11 08:24, Matt Thyer wrote: On Sep 26, 2011 2:33 PM, Adrian Chaddadr...@freebsd.org wrote: On 26 September 2011 12:56, Kevin Obermankob6...@gmail.com wrote: I suspect that the thumb drive is the issue, not FreeBSD. I've booted the drive successfully on an older Via C3 board. I'm now watching it boot successfully on a Gigabyte GA-8I915PC Duo board. I'm going to try dumping a Linux distro on this USB disk after I get PCBSD 9.0-BETA2 installed. That way I can try to eliminate the USB disk as being funny. (Although it's quite possible the USB disk is doing other funny things.) I really do think that this BIOS of mine dislikes a GPT partition inside an MBR/DOS setup. If this is the case, I wonder how many other motherboard/BIOSes have this problem. Adrian I believe this would be due to the improper GPT partition table on the FreeBSD memstick (as there is no backup partition table at the end of the volume). This was discussed earlier and now that makefs has patches for UFS label support there should be no need to continue to use GPT partitioning for the memstick image. The background is that bsdinstall relys on labels (a good thing) and GPT partition labels were being used but as there's no way to predict the size of your memstick, the resulting GPT partition table was not valid (for strict UEFI systems). The fix was to go back to traditional MS-DOS partitioning (MBR) but this couldn't be done until makefs supported UFS labels. Someone came up with patches so 9.0 should be able to have an MBR memstick and bsdinstall. I'm not sure how far this has progressed. I don't believe these patches were ever committed. Could you provide a pointer to them? -Nathan Yes, it was Andriy Gapon in this message: http://www.FreeBSD.org/cgi/mid.cgi?db=irtid=4e60f480.6040...@freebsd.org That message id link doesn't seem to work. To find the message I searched for makefs on the FreeBSD mailing list archive search facility with only Current selected. The message was dated Fri, 2nd September 2011, 18:21:36 +0300 and the link to the patch was: http://people.freebsd.org/~avg/makefs.ffs-label.diff On Sep 28, 2011 7:15 AM, Matt Thyer matt.th...@gmail.com wrote: On Sep 28, 2011 4:11 AM, Nathan Whitehorn nwhiteh...@freebsd.org wrote: On 09/27/11 08:24, Matt Thyer wrote: On Sep 26, 2011 2:33 PM, Adrian Chaddadr...@freebsd.org wrote: On 26 September 2011 12:56, Kevin Obermankob6...@gmail.com wrote: I suspect that the thumb drive is the issue, not FreeBSD. I've booted the drive successfully on an older Via C3 board. I'm now watching it boot successfully on a Gigabyte GA-8I915PC Duo board. I'm going to try dumping a Linux distro on this USB disk after I get PCBSD 9.0-BETA2 installed. That way I can try to eliminate the USB disk as being funny. (Although it's quite possible the USB disk is doing other funny things.) I really do think that this BIOS of mine dislikes a GPT partition inside an MBR/DOS setup. If this is the case, I wonder how many other motherboard/BIOSes have this problem. Adrian I believe this would be due to the improper GPT partition table on the FreeBSD memstick (as there is no backup partition table at the end of the volume). This was discussed earlier and now that makefs has patches for UFS label support there should be no need to continue to use GPT partitioning for the memstick image. The background is that bsdinstall relys on labels (a good thing) and GPT partition labels were being used but as there's no way to predict the size of your memstick, the resulting GPT partition table was not valid (for strict UEFI systems). The fix was to go back to traditional MS-DOS partitioning (MBR) but this couldn't be done until makefs supported UFS labels. Someone came up with patches so 9.0 should be able to have an MBR memstick and bsdinstall. I'm not sure how far this has progressed. I don't believe these patches were ever committed. Could you provide a pointer to them? -Nathan Yes, it was Andriy Gapon in this message: http://www.FreeBSD.org/cgi/mid.cgi?db=irtid=4e60f480.6040...@freebsd.org ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: RELENG_8 / mpt / zpool Errors
On Sep 15, 2011 11:36 PM, Tim Gustafson t...@soe.ucsc.edu wrote: The SAS 2008 chip (SAS 6G) is the one that the FreeBSD mps driver has problems with when used with port expanders. It's the older SAS 3G chip that works OK with FreeBSD I think. I had crossed some wires earlier on in our discussion. We do have a SAS 2008 chip in the system already, but it's not the one that's running the external disk boxes (and we are having no problem with those drives). The external disk boxes (the ones that are having the problems) are connect through an LSI SAS 3801E, which is handled by the mpt driver. The mpt driver is what's giving us trouble right now. Can you connect an enclosure to the SAS 2008 chip card you have now and test? I thought I had read about problems like yours with that chip on FreeBSD when using such enclosures. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: RELENG_8 / mpt / zpool Errors
On Sep 15, 2011 12:42 AM, Tim Gustafson t...@soe.ucsc.edu wrote: I don't know what's out there having chosen only one such card for home use. So you'll need to do your own research. Start with looking at any card with the right chip and then look for evidence that people have used said card with FreeBSD. LSI has the SAS 9200-8e, which is based on the SAS 2008 chipset, which the mps (not the mpt) driver claims to support. We are currently using another controller that uses the SAS 2008 chipset in this same box for the internal OS drives and have not had any problems with it. So, that makes me feel good. Does anyone have any recommendation for or against the 9200-8e? The SAS 2008 chip (SAS 6G) is the one that the FreeBSD mps driver has problems with when used with port expanders. It's the older SAS 3G chip that works OK with FreeBSD I think. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: RELENG_8 / mpt / zpool Errors
On Sep 11, 2011 6:07 AM, Tim Gustafson t...@soe.ucsc.edu wrote: I'm not sure what cards you can get now with the LSI 1068E chip. As LSI branded cards cost more, I went for for the LSI SAS2008 based Supermicro AOC-USAS2-L8i as I knew I wasn't going to use port expanders. It could be hard to get cards with the older chip now (you might have to get something second hand). I'm not at all wedded to that card; any card that has two external ports will do for me. The bigger requirement is that it works well with FreeBSD 8.2 and eventually 9.0. If you had your pick of any (reasonably priced) card that had two external ports and would work in my configuration, which would it be? I don't know what's out there having chosen only one such card for home use. So you'll need to do your own research. Start with looking at any card with the right chip and then look for evidence that people have used said card with FreeBSD. A port expander would be required and just a few older drives in the enclosure. A developer (of which I'm not) would need console access and ability to install new kernels, reboot etc. I can probably swing that, at least for a time, in 3 or 4 weeks. We have hardware on order that I can use to test with for a short while before it goes into production. You should contact the driver maintainer and see if they're available inclined to take up your offer of hardware availability. You may also have access to some talented young students who might be interested in driver programming who could help. As a production user of FreeBSD it's always great if you can find a way to contribute back to the project in areas that help you too. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: RELENG_8 / mpt / zpool Errors
On Sep 8, 2011 12:33 AM, Tim Gustafson t...@soe.ucsc.edu wrote: Advice is: Use -STABLE and not -RELEASE Upgrade firmware Avoid port multipliers As I'm using cheap 4K green drives I had to use wdidle3.exe to fix the 8 second head parking and I also had to use gnop to force my ZFS pool to use 4K transfers. We're already using -STABLE; we update to RELENG_8 periodically. LSI's web site is a little bizarre when it comes to their downloads section. I searched their drivers section for my specific model number and got a bunch of firmware upgrades for other cards with different port configurations. We can't avoid port multipliers. Nobody makes a 32-port SAS/SATA controller. And anyhow, the hardware is already purchased so I need to figure out how to make it work. :) WDC's web site says that the wdidle3.exe utility you suggested is not for this drive; the site says it's only for WD1000FYPS-01ZKB0, WD7500AYPS-01ZKB0, and WD7501AYPS-01ZKB0. I'm not sure what you mean about using gnop to force 4K transfers. The mps driver has problems with the LSI SAS2008 when using port multipliers (which are in the enclosures). If you go to the previous SAS 3G version of that chip I think you'll be OK. When you say go to the previous SAS 3G version of that chip, do you mean buy an older version of that controller card? Or are you talking about downgrading, rather than upgrading, the firmware? There are newer LSI cards out there as well. Would upgrading the card help any? The LSI SAS 9205-8e seems to be a workable solution, and not too expensive. Tim, The reference to wdidle3 is specific to my situation with the home user grade WDC20EARS drive. You have server class drives but are affected by a problem with the FreeBSD driver with port multipliers. Easiest solutions to your situation are: A) Don't use FreeBSD B) Switch to hardware that works with FreeBSD (the SAS 3G version of that LSI HBA chip) Harder solutions are: A) Fix the driver ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: RELENG_8 / mpt / zpool Errors
On Sep 9, 2011 8:38 PM, Adrian Chadd adr...@freebsd.org wrote: On 9 September 2011 18:32, Matt Thyer matt.th...@gmail.com wrote: Harder solutions are: A) Fix the driver .. which is the only way FreeBSD improves, so. You choose. :) Also: A1) Post lots of debugging info to the list, be very friendly and helpful, see if any developers are willing to help you figure and fix it A2) If not, look at paying a small (say $1k) bounty to get it fixed. :) Adrian Yes, Helping to fix the driver is a thing that Tim is in the unusual situation to be able to do as he has access to the affected hardware. However, if paid work demands a quicker solution, I'm hoping they'll fund use of the SAS 3G card (cheap) whilst allowing Tim to occasionally work with the SAS 6G card to work on a final solution! Ideally we don't want Tim to have to move away from FreeBSD. We want him to help fix the driver. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: RELENG_8 / mpt / zpool Errors
On 10 September 2011 01:19, Tim Gustafson t...@soe.ucsc.edu wrote: I can most likely switch to a different card. Just to make sure I get the right one, do you have a part number of the card that you are recommending? As I mentioned earlier, I was looking around on the LSI site and they have lots of options, but none with a port configuration that I need (2 external ports only). I'm not sure what cards you can get now with the LSI 1068E chip. As LSI branded cards cost more, I went for for the LSI SAS2008 based Supermicro AOC-USAS2-L8i as I new I wasn't going to use port expanders. It could be hard to get cards with the older chip now (you might have to get something second hand). The Supermicro AOC-USAS-L4i wouldn't be enough for you as it's only got 1 external connector. http://www.supermicro.com/products/accessories/addon/AOC-USAS-L4i_R.cfm I'd be happy to help with debugging the 6G problem. I could probably install that card into another box and give some developers access to it, but I'd have to rustle up some disks and port expanders to make it a fair test. A port expander would be required and just a few older drives in the enclosure. A developer (of which I'm not) would need console access and ability to install new kernels, reboot etc. Good luck... ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: RELENG_8 / mpt / zpool Errors
On 7 September 2011 13:54, Tim Gustafson t...@soe.ucsc.edu wrote: What are the drives exactly? You may have issues like TLER or frequent head parking. Are these SATA, SCSI or SAS and are port multipliers in use? root@bsd-03: dmesg | grep da6 da6 at mpt0 bus 0 scbus1 target 1 lun 0 da6: ATA WDC WD2003FYYS-0 1D01 Fixed Direct Access SCSI-5 device da6: 300.000MB/s transfers da6: Command Queueing enabled da6: 1907729MB (3907029168 512 byte sectors: 255H 63S/T 243201C) The drives are housed in two of the following enclosures: ses0 at mpt0 bus 0 scbus1 target 32 lun 0 ses0: PROMISE 3U-SAS-16-D BP 0107 Fixed Enclosure Services SCSI-5 device ses0: 300.000MB/s transfers ses0: Command Queueing enabled ses0: SCSI-3 SES Device Each enclosure is directly connected by a dedicated cable to the 2-port controller: mps0: LSI SAS2008 port 0xfc00-0xfcff mem 0xef5f-0xef5f,0xef58-0xef5b irq 44 at device 0.0 on pci1 mps0: Firmware: 07.15.04.00 mps0: IOCCapabilities: 185cScsiTaskFull,DiagTrace,SnapBuf,EEDP,TransRetry,IR mps0: [ITHREAD] -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Tim Gustafson t...@soe.ucsc.edu Baskin School of Engineering 831-459-5354 UC Santa Cruz Baskin Engineering 317B -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- You'll be interested in the thread PCIe SATA HBA for ZFS on -STABLE on the FreeBSD-STABLE list between 31 May 2011 and 12 June 2011. Jeremy Chadwick's post on the 1st of June is particularly enlightening. Advice is: Use -STABLE and not -RELEASE Upgrade firmware Avoid port multipliers I'm having no troubles but with 8 identical SATA disks instead of SAS. As I'm using cheap 4K green drives I had to use wdidle3.exe to fix the 8 second head parking and I also had to use gnop to force my ZFS pool to use 4K transfers. mps0: LSI SAS2008 port 0xee00-0xeeff mem 0xfbdfc000-0xfbdf,0xfbd8-0xfbdb irq 16 at device 0.0 on pci1 mps0: Firmware: 07.00.00.00 da0 at mps0 bus 0 scbus0 target 0 lun 0 da0: ATA WDC WD20EARS-00M AB51 Fixed Direct Access SCSI-5 device da0: 300.000MB/s transfers da0: Command Queueing enabled da0: 1907729MB (3907029168 512 byte sectors: 255H 63S/T 243201C) ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: RELENG_8 / mpt / zpool Errors
On 7 September 2011 19:07, Matt Thyer matt.th...@gmail.com wrote: On 7 September 2011 13:54, Tim Gustafson t...@soe.ucsc.edu wrote: What are the drives exactly? You may have issues like TLER or frequent head parking. Are these SATA, SCSI or SAS and are port multipliers in use? root@bsd-03: dmesg | grep da6 da6 at mpt0 bus 0 scbus1 target 1 lun 0 da6: ATA WDC WD2003FYYS-0 1D01 Fixed Direct Access SCSI-5 device da6: 300.000MB/s transfers da6: Command Queueing enabled da6: 1907729MB (3907029168 512 byte sectors: 255H 63S/T 243201C) The drives are housed in two of the following enclosures: ses0 at mpt0 bus 0 scbus1 target 32 lun 0 ses0: PROMISE 3U-SAS-16-D BP 0107 Fixed Enclosure Services SCSI-5 device ses0: 300.000MB/s transfers ses0: Command Queueing enabled ses0: SCSI-3 SES Device Each enclosure is directly connected by a dedicated cable to the 2-port controller: mps0: LSI SAS2008 port 0xfc00-0xfcff mem 0xef5f-0xef5f,0xef58-0xef5b irq 44 at device 0.0 on pci1 mps0: Firmware: 07.15.04.00 mps0: IOCCapabilities: 185cScsiTaskFull,DiagTrace,SnapBuf,EEDP,TransRetry,IR mps0: [ITHREAD] -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Tim Gustafson t...@soe.ucsc.edu Baskin School of Engineering 831-459-5354 UC Santa Cruz Baskin Engineering 317B -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- You'll be interested in the thread PCIe SATA HBA for ZFS on -STABLE on the FreeBSD-STABLE list between 31 May 2011 and 12 June 2011. Jeremy Chadwick's post on the 1st of June is particularly enlightening. Advice is: Use -STABLE and not -RELEASE Upgrade firmware Avoid port multipliers I'm having no troubles but with 8 identical SATA disks instead of SAS. As I'm using cheap 4K green drives I had to use wdidle3.exe to fix the 8 second head parking and I also had to use gnop to force my ZFS pool to use 4K transfers. mps0: LSI SAS2008 port 0xee00-0xeeff mem 0xfbdfc000-0xfbdf,0xfbd8-0xfbdb irq 16 at device 0.0 on pci1 mps0: Firmware: 07.00.00.00 da0 at mps0 bus 0 scbus0 target 0 lun 0 da0: ATA WDC WD20EARS-00M AB51 Fixed Direct Access SCSI-5 device da0: 300.000MB/s transfers da0: Command Queueing enabled da0: 1907729MB (3907029168 512 byte sectors: 255H 63S/T 243201C) A bit more info... The mps driver has problems with the LSI SAS2008 when using port multipliers (which are in the enclosures). If you go to the previous SAS 3G version of that chip I think you'll be OK. Either that or run Solaris Express or fix the driver. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: RELENG_8 / mpt / zpool Errors
On Sep 7, 2011 8:53 AM, Tim Gustafson t...@soe.ucsc.edu wrote: Hi all, I'm running RELENG_8: -- root@bsd-03: uname -a FreeBSD bsd-03 8.2-STABLE FreeBSD 8.2-STABLE #0: Mon Aug 22 14:58:58 PDT 2011 root@bsd-03:/usr/obj/usr/src/sys/GENERIC amd64 -- We've got an MPT controller installed with 32 drives attached: -- root@bsd-03: dmesg | grep mpt mpt0: LSILogic SAS/SATA Adapter port 0xec00-0xecff mem 0xef3fc000-0xef3f,0xef3e-0xef3e irq 32 at device 0.0 on pci3 mpt0: [ITHREAD] mpt0: MPI Version=1.5.19.0 ses0 at mpt0 bus 0 scbus1 target 32 lun 0 ses1 at mpt0 bus 0 scbus1 target 33 lun 0 da5 at mpt0 bus 0 scbus1 target 0 lun 0 .SNIP. da36 at mpt0 bus 0 scbus1 target 31 lun 0 -- We have a zpool on those drives configured into one large zfs file system: [snip] We're seeing some occasional oddness. About every two weeks it seems the controller temporarily loses connectivity with the drives and the zpool goes a bit bonkers and reports a dozen or so corrupted files. A zpool scrub goes through and reports that everything's been fixed and everything seems OK again (although I have not 100% confirmed that there is no file corruption yet, but I'm giving ZFS's check-summing logic the benefit of the doubt here). [snip] So, is this an OS/driver issue? Is it a bad controller? Bad cables? Bad disks? As always, any help is greatly appreciated. Thanks! -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Tim Gustafson t...@soe.ucsc.edu Baskin School of Engineering 831-459-5354 UC Santa Cruz Baskin Engineering 317B -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- What are the drives exactly? You may have issues like TLER or frequent head parking. Are these SATA, SCSI or SAS and are port multipliers in use? ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: http://www.freebsd.org/marketing/os-comparison.html
Advocacy by the project members is not going to be taken as seriously as an independent third party comparison. It's clear to me that the project should stick to improving it's own feature set and leave these sorts of things to others. Otherwise we're straying into Fanboy territory which aint pretty. Once we have some world beating (or even close to equalling) performance we can start to request that third parties take us seriously. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Problems booting 9.0-BETA1 memstick
Shouldn't we use MBR partitioning instead of GPT for the memstick image ? We won't need larger than 2TiB installation media for many decades! ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Problems booting 9.0-BETA1 memstick
On Sep 2, 2011 4:35 AM, Nathan Whitehorn nwhiteh...@freebsd.org wrote: On 09/01/11 14:00, Matt Thyer wrote: Shouldn't we use MBR partitioning instead of GPT for the memstick image ? We won't need larger than 2TiB installation media for many decades! It uses GPT so that the partition can be labeled, and fstab will not need e.g. da0 hard-coded into it. makefs, which builds the filesystem, does not support UFS labels. -Nathan A hack to transparently ignore GPT partition problems on the install media might be required. Or else a utility to fix the memstick partition table. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Problems booting 9.0-BETA1 memstick
On Sep 2, 2011 5:25 AM, Warren Block wbl...@wonkity.com wrote: On Thu, 1 Sep 2011, Garrett Cooper wrote: On Sep 1, 2011, at 12:00 PM, Matt Thyer wrote: Shouldn't we use MBR partitioning instead of GPT for the memstick image ? It seems wrong to have gptboot just ignore backup GPT data, but it could be an option. Can it actually do something if the backup GPT differs, or is it testing for an error that can't be handled? MBR didn't have backup data (Luxury! Pure luxury!), and it did okay. A custom hacked gptboot just for the memstick? Sorry Been working with Linux too much lately. Probably the real solution is to use MBR and have BSDinstall and the install kernel support UFS labels. Space on the memstick is not an issue. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Problems booting 9.0-BETA1 memstick
On Sep 2, 2011 7:59 AM, Nathan Whitehorn nwhiteh...@freebsd.org wrote: On 09/01/11 17:23, Matt Thyer wrote: On Sep 2, 2011 5:25 AM, Warren Blockwbl...@wonkity.com wrote: On Thu, 1 Sep 2011, Garrett Cooper wrote: On Sep 1, 2011, at 12:00 PM, Matt Thyer wrote: Shouldn't we use MBR partitioning instead of GPT for the memstick image ? It seems wrong to have gptboot just ignore backup GPT data, but it could be an option. Can it actually do something if the backup GPT differs, or is it testing for an error that can't be handled? MBR didn't have backup data (Luxury! Pure luxury!), and it did okay. A custom hacked gptboot just for the memstick? Sorry Been working with Linux too much lately. Probably the real solution is to use MBR and have BSDinstall and the install kernel support UFS labels. Space on the memstick is not an issue. Both support that. The issue is makefs(8). -Nathan During production of the memstick image? So don't use makefs then. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: Multithread Make in multicore server
On 5 October 2010 04:53, Rui Paulo rpa...@freebsd.org wrote: On 4 Oct 2010, at 18:46, Kevin Mai wrote: I see that there's no multithreading when running make.. is there a way to enable multiprocessing when running make? Try 'make -j16 buildworld'. 16 is the maximum number of parallels processes that make is going to run. Try -j32 or even -j48 and let us know which is faster. -j16 will still show lots of idle time. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: I need reply in Embedded FreeBSD Kernel Theme
On 24 June 2010 11:06, Mohammed Farrag mfar...@freebsd.org wrote: @ Matt Thanx for ur reply Matt. / FreeBSD is already a very modular system and the traditional way (a traditional way) to build for embedded systems is to follow the NanoBSD build method (tools included in the source tree) with a stripped down kernel in which you only load the modules your hardware requires using the FreeBSD loader (or after the initial boot). yeah I read about that. My mentor suggested that before and my idea is very close to NanoBSD but I don't know if the freebsd loader will load the moduls based on the hardware requirements or user requirements. I will be glad to reply me. Modules are loaded by the user adding entries to /boot/loader.conf. e.g. if_sis_load=YES. That example will load the sis driver for the Silicon Image network interfaces on my Soekris net4801 board as I have removed almost everything in my kernel and just load the modules I require. Some modules will automatically load when another driver requires them. FreeBSD does not try to discover and reconfigure your hardware at boot time like the kudzu utility in Linux. Instead it will attach to the hardware for which you have drivers in your kernel or for which you have told the loader to load modules for. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: I need reply in Embedded FreeBSD Kernel Theme
On 12 June 2010 12:02, Chargen char...@gmail.com wrote: Sir, first of all: I'm no expert in this field but I've seen your document and I'm still wondering why you should impose such a design. I suppose it's well thought of but still I'm a bit opposed to binary configuration files because I think has to be some kind of dependency on how to generate these kind of files (correct me if I'm wrong?) as far as your document goes: We will unload all the drivers that indicated with zeros in the module metadata file. That would make the OS to be a few of Megabytes. unload? what is the logic here? I'm sorry but what is the real gain here, can you please elaborate? FreeBSD is already a very modular system and the traditional way (a traditional way) to build for embedded systems is to follow the NanoBSD build method (tools included in the source tree) with a stripped down kernel in which you only load the modules your hardware requires using the FreeBSD loader (or after the initial boot). However my Soekris net4801 board still takes about 2.5 minutes to boot and I think time could be saved by parallel probing of hardware where possible. Much work has been done on fast boot times in the Linux world including an impressive demonstration by an Intel team for car instrumentation panels (on Youtube... Google for fastest Linux boot). I'd vote for more work on FreeBSD's existing boot method rather than an entirely new implementation. What problem are you trying to solve Mohammed ? ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: fsck unable to read disk sectors
It wouldn't be the BSD way to try to stop the user shooting themselves in the foot. And I agree too as it wouldn't be right for glabel to try to keep track of all possible uses for a volume and know whether each is present. That would be a typical Linux type solution. However, would it be too much for glabel to just know about UFS and tell the user to use tunefs instead if there appears to be a UFS filesystem present ? On 17 May 2010 23:32, Bernd Walter ti...@cicely7.cicely.de wrote: On Mon, May 17, 2010 at 10:54:17PM +0930, Matt Thyer wrote: On 12 May 2010 11:16, Bernd Walter ti...@cicely7.cicely.de wrote: On Tue, May 11, 2010 at 10:15:13PM +0200, Alexander Best wrote: i've posted a log here which is pretty self explanatory: http://pastebin.com/tn3NiDDW [snip] One of the typical problems users have is that they forget that adding a label takes one sector, so the labeled device is smaller. This is no problem if you create the filesystem on the labeled drive, but often enough people add the label after creating the filesystem. FreeBSD's utilities should be able to detect this situation and either correct the filesystem size or refuse to apply the label. How can this work? glabel doesn't know anything about volume contents - it just writes a label-sector and offers the remaning storage as a new volume. Result: Refusing is impossible. Changing UFS filesystem size isn't an easy task and the last sector is already lost when filesystem comes into game. Result: Too late. I think the only reasonable thing to be done is that fsck can speak up by checking the volume size with the filesystems size _after_ glabel has overwritten the last sector. -- B.Walter be...@bwct.de http://www.bwct.de Modbus/TCP Ethernet I/O Baugruppen, ARM basierte FreeBSD Rechner uvm. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
Re: fsck unable to read disk sectors
On 12 May 2010 11:16, Bernd Walter ti...@cicely7.cicely.de wrote: On Tue, May 11, 2010 at 10:15:13PM +0200, Alexander Best wrote: i've posted a log here which is pretty self explanatory: http://pastebin.com/tn3NiDDW [snip] One of the typical problems users have is that they forget that adding a label takes one sector, so the labeled device is smaller. This is no problem if you create the filesystem on the labeled drive, but often enough people add the label after creating the filesystem. FreeBSD's utilities should be able to detect this situation and either correct the filesystem size or refuse to apply the label. Everything seems to work fine until the FS decides to use that special sector. I wouldn't add a label for ufs anyway, since UFS has labeling itself, which is also handled by glabel module and doesn't require extra space. Just setup the ufs label with tunefs -L and use the resulting /dev/ufs/... device. You only need extra label for swap, but this is not problem, since it has no persistent ondisk structures. [snip] -- B.Walter be...@bwct.de http://www.bwct.de Modbus/TCP Ethernet I/O Baugruppen, ARM basierte FreeBSD Rechner uvm. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to freebsd-current-unsubscr...@freebsd.org
