Score:0

Apache 2.4 (52) debugging, several stuck apache2 -k start at 99%-ish CPU. strace stuck on FUTEX

in flag
PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND
3023 www-data  20   0  262216  19280   8848 R  93.8   0.2   3:47.05 /usr/sbin/apach

Kernel release: 5.15.0-1002-aws

There are several that get stuck this way. Most do not return anything using strace -p . I've tried the root or earliest pid, but I can't find the culprit. I've tried:

rm /dev/random 
mknod -m 644 /dev/random c 1 9 

No luck.

Sometimes I get:

db1:/root# screen
db1:/root# strace -p 3015
strace: Process 3015 attached
futex(0x7f75134a7000, FUTEX_LOCK_PI, NULL^Cstrace: Process 3015 detached
 <detached ...>

The root pid (lowest pid) says this:

db1:/var/log/apache2# strace -p 3002
strace: Process 3002 attached
pselect6(0, NULL, NULL, NULL, {tv_sec=0, tv_nsec=765119225}, NULL) = 0 (Timeout)
wait4(-1, 0x7ffdd64a2f64, WNOHANG|WSTOPPED, NULL) = 0
times({tms_utime=3 /* 0.03 s */, tms_stime=4 /* 0.04 s */, tms_cutime=0, tms_cstime=0}) = 1718048747
pselect6(0, NULL, NULL, NULL, {tv_sec=1, tv_nsec=0}, NULL) = 0 (Timeout)
wait4(-1, 0x7ffdd64a2f64, WNOHANG|WSTOPPED, NULL) = 0

I've updated, rebooted, it is a new installation of Ubuntu. Not sure what is wrong!

Nothing present in the apache logs or system logs. I can't find a specific error or error that would be of relation.

mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.