Re: help with crash dump

2019-09-24 Thread Igor V. Ruzanov
What kernel modules do you load with this kernel? (show the list via kldstat) |Randomly every few weeks / months one of the servers I manage gets a panic. It doesn't happen on the same server every time. | |I run FreeBSD stable compiled on 2 September 2019 but it did happen with previous versi

Re: help with crash dump

2019-09-24 Thread Christos Chatzaras
Id Refs AddressSize Name 1 12 0x8020 2446480 kernel 21 0x82648000281e0 geom_mirror.ko 31 0x82a21000 88e8 tmpfs.ko 41 0x82a2a000258f8 ipfw.ko 51 0x82a5 acf mac_ntpd.ko 61 0x82a51000

Re: help with crash dump

2019-09-24 Thread Igor V. Ruzanov
Is your kernel custom and you were compile it? If yes, could you please send kernel config to here? |Id Refs AddressSize Name | 1 12 0x8020 2446480 kernel | 21 0x82648000281e0 geom_mirror.ko | 31 0x82a21000 88e8 tmpfs.ko | 41 0xff

Re: help with crash dump

2019-09-24 Thread Christos Chatzaras
> On 24 Sep 2019, at 12:43, Igor V. Ruzanov wrote: > > Is your kernel custom and you were compile it? If yes, could you please > send kernel config to here? The kernel is from 12.0-STABLE r351639. I use the GENERIC config to compile the kernel but with custom src.conf: cat /etc/src.conf W

Re: help with crash dump

2019-09-24 Thread Igor V. Ruzanov
Please try to remove 'WITHOUT_REPRODUCIBLE_BUILD' option and recompile the kernel to see if it will be stable in your context. The modern FreeBSD releses (12.0 for example) is with reproducible build enabled now. As quck check, tell uname -a and you will give the following output: FreeBSD xxx.yo

Question about bottle neck in storage

2019-09-24 Thread John Fleming
Is there anyway to see how busy a SAS/Sata controller is vs disks? I have a R720 with 14 Samsung 860 EVOs in it (its a lab server) in raid 10 ZFS. When firing off a dd I (bs=1G count=10) seems like the disks never go above %50 busy. I'm trying to figure out if i'm maxing out SATA 3 BW or if its so

Re: Question about bottle neck in storage

2019-09-24 Thread Pete Wright
On 9/24/19 8:45 AM, John Fleming wrote: Is there anyway to see how busy a SAS/Sata controller is vs disks? I have a R720 with 14 Samsung 860 EVOs in it (its a lab server) in raid 10 ZFS. When firing off a dd I (bs=1G count=10) seems like the disks never go above %50 busy. I'm trying to figure

Re: Question about bottle neck in storage

2019-09-24 Thread John Fleming
On Tue, Sep 24, 2019 at 12:05 PM Pete Wright wrote: > > > > On 9/24/19 8:45 AM, John Fleming wrote: > > Is there anyway to see how busy a SAS/Sata controller is vs disks? I > > have a R720 with 14 Samsung 860 EVOs in it (its a lab server) in raid > > 10 ZFS. > > > > When firing off a dd I (bs=1G c

Re: Question about bottle neck in storage

2019-09-24 Thread Richard Mackerras
I’d be interested to know what the actual throughput is you are getting. Are the disks SATA 7200RPM? What speed is the disk interface? Do you have just 2 disks or more? Richard [Richard Mackerras - Chat @ Spike](https://www.spikenow.com/?ref=spike-organic-signature&_ts=6bswo) [6bswo] On Se

Re: Question about bottle neck in storage

2019-09-24 Thread Warner Losh
On Tue, Sep 24, 2019 at 5:46 PM John Fleming wrote: > Is there anyway to see how busy a SAS/Sata controller is vs disks? I > have a R720 with 14 Samsung 860 EVOs in it (its a lab server) in raid > 10 ZFS. > > When firing off a dd I (bs=1G count=10) seems like the disks never go > above %50 busy.

Re: Missing 12 MFCs

2019-09-24 Thread Ed Maste
On Thu, 19 Sep 2019 at 18:30, David Cross wrote: > > There are a set of commits in head that have not been pulled into the > stable/12 branch. Many of these are marked as MFC (and are quite old). I'm looking at the WITH_PIE / WITH_BIND_NOW changes, and am looking for re@'s opinion. I'll merge th