Comment 78 for bug 1711407

Hi Dan,

the CPU usage is zero, there is nothing popping up. The tasks causing the high load value are in D state. If I do "echo w > /proc/sysrq-trigger", I will get for every leaked NS:

[392149.562095] kworker/u81:40 D 0 4290 2 0x80000000
[392149.562112] Workqueue: netns cleanup_net
[392149.562125] Call Trace:
[392149.562135] __schedule+0x291/0x8a0
[392149.562146] schedule+0x2c/0x80
[392149.562157] schedule_timeout+0x182/0x360
[392149.562167] ? call_timer_fn+0x130/0x130
[392149.562177] msleep+0x2d/0x40
[392149.562188] ? msleep+0x2d/0x40
[392149.562199] netdev_run_todo+0x15e/0x5d0
[392149.562210] rtnl_unlock+0xe/0x10
[392149.562221] default_device_exit_batch+0x156/0x180
[392149.562233] ? do_wait_intr_irq+0x90/0x90
[392149.562245] cleanup_net+0x14e/0x2a0
[392149.562257] process_one_work+0x1ec/0x410
[392149.562270] worker_thread+0x32/0x410
[392149.562281] kthread+0x128/0x140
[392149.562292] ? process_one_work+0x410/0x410
[392149.562300] ? kthread_create_on_node+0x70/0x70
[392149.562305] ? do_syscall_64+0x67/0x130
[392149.562310] ? SyS_exit_group+0x14/0x20
[392149.562316] ret_from_fork+0x35/0x40

I am not sure how fine is to have e.g. thousands of such leaked namespaces - if I can't run out of some resources (kworker threads?). I am going to keep the machine running + generate even higher load there to see if I hit any such limit.

Thanks
JH