Jump to content
funtoo forums

vkraven

Members
  • Content Count

    4
  • Joined

  • Last visited

  • Days Won

    1

Everything posted by vkraven

  1. Hey everyone. I'd been having issues importing zpools in ZFS since a recent update (I can't recall for sure, maybe 0.6.4 or 0.6.5). For the longest time, I thought I hadn't compiled my kernel properly, but this helped me. https://github.com/zfsonlinux/zfs/issues/4337 It seems this issue occurs for gentoo-patched ZFS, and the poster worked around it by compiling ZOL from the ZOL Project's tarball. Having done the same, I can attest that this worked for me too. Something about the gentoo patches might be messing with it. Although it's a gentoo package - considering this compiling from source is just an ad hoc workaround, I thought you guys would like to know. Hope it helps.
  2. I like iwoloschin's idea, because dd does a "proper" clone of your data. Some notes from my experience - 1) dd will copy every bit, even free space. So your old 100GB partition of which you have used 50GB will still create a 100GB img. cp, on the other hand, will only take note of the files you have used. 2) As such, you won't be able to use dd if=/dev/sdX of=/dev/sdY if your destination drive is even slightly smaller than your original. You could try it with partitions, though! 3) I suppose, to get around the partition resizing issue, you could partition first and then use dd on each partition, as long as the size rule in 2) is met. 4) dd takes a really long time and has no progress bar. So I'd advise patience, free time and planning ahead.
  3. Some googling tells me that's a Sandy Bridge CPU. I'm not completely certain, however, if intel64-sandybridge will work since the wiki page does say Core i3/5/7, although I can't think of any reason why it shouldn't. Should you want to play it a little safer though, you could go with corei7.
  4. Wasn't sure if I should have posted this in "Help Central" or here, but hopefully this will be useful for someone facing similar problems. I installed Funtoo on a Samsung 850 Pro 256GB SSD recently. All went well, but I noticed that, while using gnome 3.14.2, certain apps would trigger a gradual desktop freeze (google chrome). Chrome would freeze, then terminal, then overview - but the mouse cursor remained responsive. After a minute or so, everything would work again. A bit of googling led me to check dmesg, which produced the errors: ata1.00: failed command: WRITE FPDMA QUEUED ata1.00: cmd 61/00:f0:10:3b:67/02:00:2a:00:00/40 tag 30 ncq 262144 out res 40/00:01:00:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout) ata1.00: status: { DRDY } ata1: hard resetting link ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300) ata1.00: supports DRM functions and may not be fully accessible ata1.00: device reported invalid CHS sector 0 ata1.00: device reported invalid CHS sector 0 ata1.00: device reported invalid CHS sector 0 ata1.00: device reported invalid CHS sector 0 etc, etc, over and over Which I eventually figured out was a problem with NCQ not playing well with the 850 Pro. What I found curious was that not all the 850 Pros had this issue. Apparently, in addition to the controversial dead after firmware update troubles (http://forums.anandtech.com/showthread.php?t=2421585), this error seems to primarily affect the new firmware EXM02B6Q as well - http://www.spinics.net/lists/linux-ide/msg50342.html - consequently affecting my SSD, which shipped with the new firmware. I got around it by disabling NCQ like so - echo 1 > /sys/block/sdX/device/queue_depth for whatever X is the 850 Pro drive. E.g. for me, echo 1 > /sys/block/sda/device/queue_depth But this workaround does not persist through boots, and I don't understand code well enough to program a startup script. Since this NCQ error did occur for me during bootup as well (where my screen would stay black for a minute before gnome loads), I added libata.force=noncq into Funtoo's boot configs, like so: nano -w /etc/boot.conf "Funtoo Linux genkernel" { kernel kernel[-v] initrd initramfs[-v] params += real_root=auto libata.force=noncq } And then ran boot-update The only drawback I can think of to this is that doing so would disable NCQ for all my other drives which do not have NCQ issues, as well. Hopefully this will help anyone facing these annoying freezes on one of Samsung's drives too.
  5. I got that issue today too. Was this other package jsoncpp? I got around it by emerging an older version of cmake, cmake-2.8.12.2, then emerging jsoncpp, and updating cmake to current. To have portage emerge the older version, try echo ">dev-util/cmake-2.8.12.2" >> /etc/portage/package.mask Once that's done, I removed the line from package.mask Hope it works for you!
×
×
  • Create New...