Public issue detail

Runaway CPU investigation for sshd-auth: timer churn at read_tsc

sshd-auth is stuck in a likely timer churn loop: 50.00% of sampled CPU passed through read_tsc, with repeated read x3, write x2, clock_nanosleep x1.

highpackage: openssh-serversource: opensshscore: 106reports: 1

Last seen: 2026-06-28 11:24 UTC. Public JSON: /v1/issues/019f0b26-eaa6-7da3-8004-93f5ebf26a41

Technical snapshot

This is the clearest retained userspace thread cluster Fixer captured while the process was spinning.

Representative thread backtrace

  • Command: sshd-auth: root [net]
  • Why Fixer classified it this way: Timer-related syscalls or symbols dominate the sample, which suggests a wakeup timer is firing too aggressively.
  • Wait site: unix_stream_read_generic
  • Hot path: read_tsc (50.00% sampled CPU)
  • Repeated loop: clock_nanosleep -> getrandom -> write
  • Top syscalls: read x3, write x2, clock_nanosleep x1, exit_group x1
  • Package: openssh-server 1:10.3p1-1
  • Kernel: 6.17.10+deb14-amd64
  • Distribution: debian
unix_stream_read_generic
unix_stream_recvmsg
sock_recvmsg
sock_read_iter
vfs_read
ksys_read
do_syscall_64
entry_SYSCALL_64_after_hwframe

Possible duplicates

These are suggestions based on sanitized trigram similarity plus structured fields like package, subsystem, classification, and wait site. They are not auto-merged.

sshd-auth is stuck in a likely timer churn loop: 100.00% of sampled CPU passed through apparmor_socket_recvmsg, with repeated thread backtraces show 1 thread(s) around 0x00007fa460a9a7d2 in ?? () from target:<path>.

highpackage: openssh-serversource: opensshscore: 106reports: 2patch attempt readysimilarity: 99%

Why this looks related: same classification, same package, same source package, same subsystem, same target, same wait site

Last seen: 2026-07-10 03:06 UTC. Public page: /issues/019dfcf9-7385-7c01-ba59-9fce98cd0a8f. Public JSON: /v1/issues/019dfcf9-7385-7c01-ba59-9fce98cd0a8f

sshd-auth is stuck in a likely timer churn loop: 100.00% of sampled CPU passed through vfs_write, with repeated thread backtraces show 1 thread(s) around 0x00007f76d5c9a7d2 in ?? () from target:<path>.

highpackage: openssh-serversource: opensshscore: 106reports: 2patch attempt readysimilarity: 99%

Why this looks related: same classification, same package, same source package, same subsystem, same target, same wait site

Last seen: 2026-07-10 03:06 UTC. Public page: /issues/019e2c59-6707-7471-b7b2-5329d3cdd34e. Public JSON: /v1/issues/019e2c59-6707-7471-b7b2-5329d3cdd34e

sshd-auth is stuck in a likely busy poll loop: 100.00% of sampled CPU passed through rb_erase, with repeated read x3, write x2, clock_nanosleep x1.

highpackage: openssh-serversource: opensshscore: 106reports: 1similarity: 99%

Why this looks related: same package, same source package, same subsystem, same target, same wait site

Last seen: 2026-07-08 12:09 UTC. Public page: /issues/019f2502-84e9-7630-abd1-0bd0d62fcc9c. Public JSON: /v1/issues/019f2502-84e9-7630-abd1-0bd0d62fcc9c

sshd-auth is stuck in a likely busy poll loop: 100.00% of sampled CPU passed through unresolved offset, with repeated read x3, write x2, clock_nanosleep x1.

highpackage: openssh-serversource: opensshscore: 106reports: 2patch attempt readysimilarity: 94%

Why this looks related: same package, same source package, same subsystem, same target, same wait site

Last seen: 2026-07-10 03:06 UTC. Public page: /issues/019e1519-78c6-7813-bebb-aeb6177e1897. Public JSON: /v1/issues/019e1519-78c6-7813-bebb-aeb6177e1897

sshd-auth is stuck in a likely unclassified userspace loop: 50.00% of sampled CPU passed through schedule_timeout, with repeated restart_syscall x1.

highpackage: openssh-serversource: opensshscore: 106reports: 1similarity: 75%

Why this looks related: same package, same source package, same subsystem, same target

Last seen: 2026-07-02 22:27 UTC. Public page: /issues/019f0e0a-9b12-7bc3-ba1f-6777e4555f8d. Public JSON: /v1/issues/019f0e0a-9b12-7bc3-ba1f-6777e4555f8d

sshd-auth is stuck in a likely unclassified userspace loop: 50.00% of sampled CPU passed through fdget_pos, with repeated no dominant syscall sample was captured.

highpackage: openssh-serversource: opensshscore: 106reports: 1similarity: 72%

Why this looks related: same package, same source package, same subsystem, same target

Last seen: 2026-06-27 21:56 UTC. Public page: /issues/019f0a5d-c564-7ca1-bf39-b796c2e8376d. Public JSON: /v1/issues/019f0a5d-c564-7ca1-bf39-b796c2e8376d

Published attempts

No public attempts have been published for this issue yet.