On Fri, 14 Aug 2015, Tim Daneliuk wrote:> I just built a 10.2 machine on a cloud-based VPS (Digital Ocean) that has > 512M of memory and 1G of swap partition. I am seeing a ton of errors like > this: > > Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(10): failed > Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(14): failed > Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(11): failed > Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(6): failed > Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(7): failed > Aug 14 00:01:22 myhost last message repeated 2 times > > > So, I added this to fstab (after creating /usr/swap0): > > md99 none swap sw,file=/usr/swap0 0 0 > > And then did this: > > swapon -aq > > > But, when I do a swapinfo, all I can see is the "disk" swap partition > that comes standard with the VPS: > > > Device 1K-blocks Used Avail Capacity > /dev/gpt/swapfs 1048576 456572 592004 44%Add the -L (late) option to swapon. How this works might differ between 10-Release, 10-Stable, and 11. Incidentally, md99 does not have to be literal, it's just meant to get the md device number up out of the way of common interactive usage of mdconfig.
On 08/14/2015 12:39 PM, Warren Block wrote:> On Fri, 14 Aug 2015, Tim Daneliuk wrote: > >> I just built a 10.2 machine on a cloud-based VPS (Digital Ocean) that has >> 512M of memory and 1G of swap partition. I am seeing a ton of errors like >> this: >> >> Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(10): failed >> Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(14): failed >> Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(11): failed >> Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(6): failed >> Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(7): failed >> Aug 14 00:01:22 myhost last message repeated 2 times >> >> >> So, I added this to fstab (after creating /usr/swap0): >> >> md99 none swap sw,file=/usr/swap0 0 0 >> >> And then did this: >> >> swapon -aq >> >> >> But, when I do a swapinfo, all I can see is the "disk" swap partition >> that comes standard with the VPS: >> >> >> Device 1K-blocks Used Avail Capacity >> /dev/gpt/swapfs 1048576 456572 592004 44% > > Add the -L (late) option to swapon. How this works might differ between 10-Release, 10-Stable, and 11. > > Incidentally, md99 does not have to be literal, it's just meant to get the md device number up out of the way of common interactive usage of mdconfig.I can try that, but this still does not resolve the issue of md99 (I get that its not literal :) is not destroyed with the swapoff nor the fact that the swap file doesn't get put into use at all on boot. -- ---------------------------------------------------------------------------- Tim Daneliuk tundra at tundraware.com PGP Key: http://www.tundraware.com/PGP/
On 08/14/2015 12:39 PM, Warren Block wrote:> On Fri, 14 Aug 2015, Tim Daneliuk wrote: > >> I just built a 10.2 machine on a cloud-based VPS (Digital Ocean) that has >> 512M of memory and 1G of swap partition. I am seeing a ton of errors like >> this: >> >> Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(10): failed >> Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(14): failed >> Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(11): failed >> Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(6): failed >> Aug 14 00:01:22 myhost kernel: swap_pager_getswapspace(7): failed >> Aug 14 00:01:22 myhost last message repeated 2 times >> >> >> So, I added this to fstab (after creating /usr/swap0): >> >> md99 none swap sw,file=/usr/swap0 0 0 >> >> And then did this: >> >> swapon -aq >> >> >> But, when I do a swapinfo, all I can see is the "disk" swap partition >> that comes standard with the VPS: >> >> >> Device 1K-blocks Used Avail Capacity >> /dev/gpt/swapfs 1048576 456572 592004 44% > > Add the -L (late) option to swapon. How this works might differ between 10-Release, 10-Stable, and 11. > > Incidentally, md99 does not have to be literal, it's just meant to get the md device number up out of the way of common interactive usage of mdconfig.So -L does fix the problem - sort of. The machine picks up the file as additional swap on boot just fine. HWOEVER, when I try to reboot or shut down the host, I get a panic telling me some noise about not being able to shutdown swap for some reason. So ... I decided to just add a second disk partition for swap and - for some reason - it works fine interactively, but upon reboot, the newly created swap partition no longer exists and gpart shows the space as free again. I tried a gpart commit, but get "operation not permitted". So now I am trying to figure out how to make gpart changes stick. This may be an artifact of the way Digital Ocean droplets are set up .... Grrrrrrrr -- ---------------------------------------------------------------------------- Tim Daneliuk tundra at tundraware.com PGP Key: http://www.tundraware.com/PGP/