[Kernel-packages] [Bug 1953058] Re: Kernel "BUG: soft lockup" with 5.4 kernels on arm64 node appleton node (dmesg spammed with "mlx5_core 0005:01:00.0: mlx5_eq_comp_int:159:(pid 1180): Completion even

2023-04-21 Thread Ike Panhc
Test from 5.4.0-26.30 and looks like this issue starts from 5.4.0-31.35. I will do more test to make sure this -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1953058 Title: Kernel "BUG:

[Kernel-packages] [Bug 1953058] Re: Kernel "BUG: soft lockup" with 5.4 kernels on arm64 node appleton node (dmesg spammed with "mlx5_core 0005:01:00.0: mlx5_eq_comp_int:159:(pid 1180): Completion even

2023-03-07 Thread Ike Panhc
Same test in #7 for bionic and so far all 13 deploy looks good, no soft lockup. Looks this is a focal kernel issue and I will try to reboot into different focal kernel. ** Changed in: linux (Ubuntu Focal) Status: Confirmed => In Progress -- You received this bug notification because you

[Kernel-packages] [Bug 1953058] Re: Kernel "BUG: soft lockup" with 5.4 kernels on arm64 node appleton node (dmesg spammed with "mlx5_core 0005:01:00.0: mlx5_eq_comp_int:159:(pid 1180): Completion even

2023-03-06 Thread Ike Panhc
Tried to deploy and wait for 100min to see if soft lockup shows. Deploy focal and I can reproduce 5 times in 8 deploy test. Deploy jammy and it passes 20 deploy and everything looks good. It looks more and more like a focal kernel issue to me. -- You received this bug notification because you

[Kernel-packages] [Bug 1953058] Re: Kernel "BUG: soft lockup" with 5.4 kernels on arm64 node appleton node (dmesg spammed with "mlx5_core 0005:01:00.0: mlx5_eq_comp_int:159:(pid 1180): Completion even

2023-01-02 Thread Ike Panhc
For next steps, 1) Find out why I can not use maas-cli to deploy bionic-hwe on d05-3 2) Collect failure logs on appleton 3) Find out the hardware difference -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu.

[Kernel-packages] [Bug 1953058] Re: Kernel "BUG: soft lockup" with 5.4 kernels on arm64 node appleton node (dmesg spammed with "mlx5_core 0005:01:00.0: mlx5_eq_comp_int:159:(pid 1180): Completion even

2023-01-02 Thread Ike Panhc
MAAS deploy/release loop with focal[1] on d05-3 and has deployed for 82 times without failure. MAAS deploy/release loop with bionic-hwe on appleton run 100 times and 10 of them are failed. Look like this issue is only happened on appleton. -- [1] For some reason I can not deploy bionic-hwe

[Kernel-packages] [Bug 1953058] Re: Kernel "BUG: soft lockup" with 5.4 kernels on arm64 node appleton node (dmesg spammed with "mlx5_core 0005:01:00.0: mlx5_eq_comp_int:159:(pid 1180): Completion even

2022-12-21 Thread Ike Panhc
Check log again and it looks like failure happens when reboot from deploying. I will try deploy/release cycle again and see how it goes. -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu.

[Kernel-packages] [Bug 1953058] Re: Kernel "BUG: soft lockup" with 5.4 kernels on arm64 node appleton node (dmesg spammed with "mlx5_core 0005:01:00.0: mlx5_eq_comp_int:159:(pid 1180): Completion even

2022-12-18 Thread Ike Panhc
appleton with 5.4.0-135.152~18.04.2-generic passes 1000 reboot without any soft lockup. I will try 5.4.0-92.103~18.04.2. d05-3 with 5.4.0-135.152~18.04.2-generic passes 669 reboot without any error. ** Changed in: linux (Ubuntu Focal) Assignee: (unassigned) => Ike Panhc (ikepanhc) -- You

[Kernel-packages] [Bug 1953058] Re: Kernel "BUG: soft lockup" with 5.4 kernels on arm64 node appleton node (dmesg spammed with "mlx5_core 0005:01:00.0: mlx5_eq_comp_int:159:(pid 1180): Completion even

2022-12-15 Thread Ike Panhc
I find 2 systems with same Mellanox NIC card and put both systems in reboot test overnight. 0005:01:00.0 Ethernet controller [0200]: Mellanox Technologies MT27710 Family [ConnectX-4 Lx] [15b3:1015] 0005:01:00.1 Ethernet controller [0200]: Mellanox Technologies MT27710 Family [ConnectX-4 Lx]

[Kernel-packages] [Bug 1953058] Re: Kernel "BUG: soft lockup" with 5.4 kernels on arm64 node appleton node (dmesg spammed with "mlx5_core 0005:01:00.0: mlx5_eq_comp_int:159:(pid 1180): Completion even

2022-12-14 Thread Ike Panhc
@cypressyew, I use another machine with mlx5 NIC but can not reproduce. I might need to borrow appleton for testing. -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1953058 Title:

[Kernel-packages] [Bug 1953058] Re: Kernel "BUG: soft lockup" with 5.4 kernels on arm64 node appleton node (dmesg spammed with "mlx5_core 0005:01:00.0: mlx5_eq_comp_int:159:(pid 1180): Completion even

2022-09-05 Thread Po-Hsu Lin
** Summary changed: - Kernel "BUG: soft lockup" with 5.4 kernels on appleton node (arm64) + Kernel "BUG: soft lockup" with 5.4 kernels on arm64 node appleton node (dmesg spammed with "mlx5_core 0005:01:00.0: mlx5_eq_comp_int:159:(pid 1180): Completion event for bogus CQ 0x5a5aa9") ** Tags