sshd high cpu : system freeze

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

sshd high cpu : system freeze

Thuban
While transferring files over SSH (either scp or sftp used) from/to an USB drive
plugged in a ROCK64 from/to another machine, the CPU usage keep increasing and
finally, the ROCK64 freeze.
Because of the freeze, I'm not sure how to get errors. Please telle me how to
debug this.

Some details :
- Both systems (server and client) are on OpenBSD : client is 6.4, ROCK64 is
last -current.
- The usb drive is mounted with this fstab entry :
        782f1ddb783cdd13.b /mnt/bigstorage ffs rw,noexec,nodev,nosuid 1 2
- The issue only happens when transferring from/to the USB hard drive.
- dmesg :

        OpenBSD 6.4-current (GENERIC.MP) #343: Mon Dec 31 18:12:02 MST 2018
                [hidden email]:/usr/src/sys/arch/arm64/compile/GENERIC.MP
        real mem  = 4213874688 (4018MB)
        avail mem = 4050837504 (3863MB)
        mainbus0 at root: Pine64 Rock64
        cpu0 at mainbus0 mpidr 0: ARM Cortex-A53 r0p4
        cpu0: 32KB 64b/line 2-way L1 VIPT I-cache, 32KB 64b/line 4-way L1 D-cache
        cpu0: 256KB 64b/line 16-way L2 cache
        efi0 at mainbus0: UEFI 2.0.5
        efi0: Das U-boot rev 0x0
        psci0 at mainbus0: PSCI 1.0
        syscon0 at mainbus0: "syscon"
        rkclock0 at mainbus0
        syscon1 at mainbus0: "syscon"
        ampintc0 at mainbus0 nirq 160, ncpu 4 ipi: 0, 1: "interrupt-controller"
        rkpinctrl0 at mainbus0: "pinctrl"
        rkgpio0 at rkpinctrl0
        rkgpio1 at rkpinctrl0
        rkgpio2 at rkpinctrl0
        rkgpio3 at rkpinctrl0
        simplebus0 at mainbus0: "amba"
        agtimer0 at mainbus0: tick rate 24000 KHz
        com0 at mainbus0: ns16550, no working fifo
        com0: console
        rkiic0 at mainbus0
        iic0 at rkiic0
        rkpmic0 at iic0 addr 0x18: RK805
        rktemp0 at mainbus0
        dwmmc0 at mainbus0: 49 MHz base clock
        sdmmc0 at dwmmc0: 4-bit, sd high-speed, mmc high-speed, dma
        dwmmc1 at mainbus0: 49 MHz base clock
        sdmmc1 at dwmmc1: 8-bit, mmc high-speed, dma
        dwge0 at mainbus0
        dwge0: address: d6:8f:58:d2:50:ce
        rgephy0 at dwge0 phy 0: RTL8169S/8110S/8211 PHY, rev. 6
        ehci0 at mainbus0
        usb0 at ehci0: USB revision 2.0
        uhub0 at usb0 configuration 1 interface 0 "Generic EHCI root hub" rev 2.00/1.00 addr 1
        cpu1 at mainbus0 mpidr 1: ARM Cortex-A53 r0p4
        cpu1: 32KB 64b/line 2-way L1 VIPT I-cache, 32KB 64b/line 4-way L1 D-cache
        cpu1: 256KB 64b/line 16-way L2 cache
        cpu2 at mainbus0 mpidr 2: ARM Cortex-A53 r0p4
        cpu2: 32KB 64b/line 2-way L1 VIPT I-cache, 32KB 64b/line 4-way L1 D-cache
        cpu2: 256KB 64b/line 16-way L2 cache
        cpu3 at mainbus0 mpidr 3: ARM Cortex-A53 r0p4
        cpu3: 32KB 64b/line 2-way L1 VIPT I-cache, 32KB 64b/line 4-way L1 D-cache
        cpu3: 256KB 64b/line 16-way L2 cache
        scsibus0 at sdmmc1: 2 targets, initiator 0
        sd0 at scsibus0 targ 1 lun 0: <SD/MMC, NCard, 0000> SCSI2 0/direct removable
        sd0: 29600MB, 512 bytes/sector, 60620800 sectors
        scsibus1 at sdmmc0: 2 targets, initiator 0
        sd1 at scsibus1 targ 1 lun 0: <SD/MMC, USD00, 0010> SCSI2 0/direct removable
        sd1: 15080MB, 512 bytes/sector, 30883840 sectors
        umass0 at uhub0 port 1 configuration 1 interface 0 "Western Digital Ext HDD 1021" rev 2.00/20.21 addr 2
        umass0: using SCSI over Bulk-Only
        scsibus2 at umass0: 2 targets, initiator 0
        sd2 at scsibus2 targ 1 lun 0: <WD, Ext HDD 1021, 2021> SCSI2 0/direct fixed serial.10581021383235373034
        sd2: 1907727MB, 512 bytes/sector, 3907024896 sectors
        vscsi0 at root
        scsibus3 at vscsi0: 256 targets
        softraid0 at root
        scsibus4 at softraid0: 256 targets
        bootfile: sd0a:/bsd
        boot device: sd0
        root on sd0a (532ea4ce9a5e001c.a) swap on sd0b dump on sd0b
        WARNING: / was not properly unmounted






--
    thuban

Reply | Threaded
Open this post in threaded view
|

High USB i/o : system freeze

Thuban
After some investigation, it seems the issue appears when there is high
i/o on an USB drive, so nothing related with SSH (I just use SSH to copy
files).

Some others seems to have the same issue under Linux with the ROCK64 :

* https://forum.pine64.org/showthread.php?tid=5137
* https://github.com/ayufan-rock64/linux-build/issues/112

On the first link, they talk about disabling UAS.

Do you have any advice to do so under OpenBSD?

Or any other advice now I think I get where is the problem ?

Regards.

* Thuban <[hidden email]> le [06-01-2019 10:51:03 +0100]:

> While transferring files over SSH (either scp or sftp used) from/to an USB drive
> plugged in a ROCK64 from/to another machine, the CPU usage keep increasing and
> finally, the ROCK64 freeze.
> Because of the freeze, I'm not sure how to get errors. Please telle me how to
> debug this.
>
> Some details :
> - Both systems (server and client) are on OpenBSD : client is 6.4, ROCK64 is
> last -current.
> - The usb drive is mounted with this fstab entry :
> 782f1ddb783cdd13.b /mnt/bigstorage ffs rw,noexec,nodev,nosuid 1 2
> - The issue only happens when transferring from/to the USB hard drive.
> - dmesg :
>
> OpenBSD 6.4-current (GENERIC.MP) #343: Mon Dec 31 18:12:02 MST 2018
> [hidden email]:/usr/src/sys/arch/arm64/compile/GENERIC.MP
> real mem  = 4213874688 (4018MB)
> avail mem = 4050837504 (3863MB)
> mainbus0 at root: Pine64 Rock64
> cpu0 at mainbus0 mpidr 0: ARM Cortex-A53 r0p4
> cpu0: 32KB 64b/line 2-way L1 VIPT I-cache, 32KB 64b/line 4-way L1 D-cache
> cpu0: 256KB 64b/line 16-way L2 cache
> efi0 at mainbus0: UEFI 2.0.5
> efi0: Das U-boot rev 0x0
> psci0 at mainbus0: PSCI 1.0
> syscon0 at mainbus0: "syscon"
> rkclock0 at mainbus0
> syscon1 at mainbus0: "syscon"
> ampintc0 at mainbus0 nirq 160, ncpu 4 ipi: 0, 1: "interrupt-controller"
> rkpinctrl0 at mainbus0: "pinctrl"
> rkgpio0 at rkpinctrl0
> rkgpio1 at rkpinctrl0
> rkgpio2 at rkpinctrl0
> rkgpio3 at rkpinctrl0
> simplebus0 at mainbus0: "amba"
> agtimer0 at mainbus0: tick rate 24000 KHz
> com0 at mainbus0: ns16550, no working fifo
> com0: console
> rkiic0 at mainbus0
> iic0 at rkiic0
> rkpmic0 at iic0 addr 0x18: RK805
> rktemp0 at mainbus0
> dwmmc0 at mainbus0: 49 MHz base clock
> sdmmc0 at dwmmc0: 4-bit, sd high-speed, mmc high-speed, dma
> dwmmc1 at mainbus0: 49 MHz base clock
> sdmmc1 at dwmmc1: 8-bit, mmc high-speed, dma
> dwge0 at mainbus0
> dwge0: address: d6:8f:58:d2:50:ce
> rgephy0 at dwge0 phy 0: RTL8169S/8110S/8211 PHY, rev. 6
> ehci0 at mainbus0
> usb0 at ehci0: USB revision 2.0
> uhub0 at usb0 configuration 1 interface 0 "Generic EHCI root hub" rev 2.00/1.00 addr 1
> cpu1 at mainbus0 mpidr 1: ARM Cortex-A53 r0p4
> cpu1: 32KB 64b/line 2-way L1 VIPT I-cache, 32KB 64b/line 4-way L1 D-cache
> cpu1: 256KB 64b/line 16-way L2 cache
> cpu2 at mainbus0 mpidr 2: ARM Cortex-A53 r0p4
> cpu2: 32KB 64b/line 2-way L1 VIPT I-cache, 32KB 64b/line 4-way L1 D-cache
> cpu2: 256KB 64b/line 16-way L2 cache
> cpu3 at mainbus0 mpidr 3: ARM Cortex-A53 r0p4
> cpu3: 32KB 64b/line 2-way L1 VIPT I-cache, 32KB 64b/line 4-way L1 D-cache
> cpu3: 256KB 64b/line 16-way L2 cache
> scsibus0 at sdmmc1: 2 targets, initiator 0
> sd0 at scsibus0 targ 1 lun 0: <SD/MMC, NCard, 0000> SCSI2 0/direct removable
> sd0: 29600MB, 512 bytes/sector, 60620800 sectors
> scsibus1 at sdmmc0: 2 targets, initiator 0
> sd1 at scsibus1 targ 1 lun 0: <SD/MMC, USD00, 0010> SCSI2 0/direct removable
> sd1: 15080MB, 512 bytes/sector, 30883840 sectors
> umass0 at uhub0 port 1 configuration 1 interface 0 "Western Digital Ext HDD 1021" rev 2.00/20.21 addr 2
> umass0: using SCSI over Bulk-Only
> scsibus2 at umass0: 2 targets, initiator 0
> sd2 at scsibus2 targ 1 lun 0: <WD, Ext HDD 1021, 2021> SCSI2 0/direct fixed serial.10581021383235373034
> sd2: 1907727MB, 512 bytes/sector, 3907024896 sectors
> vscsi0 at root
> scsibus3 at vscsi0: 256 targets
> softraid0 at root
> scsibus4 at softraid0: 256 targets
> bootfile: sd0a:/bsd
> boot device: sd0
> root on sd0a (532ea4ce9a5e001c.a) swap on sd0b dump on sd0b
> WARNING: / was not properly unmounted
>
>
>
>
>
>
> --
>     thuban
>

--
    thuban