As suggested by Chris, here the statistics:
version crash-tcnative non-crash-jsse non-crash-tcnative
11.0.29 4/308 1/77 0/308
11.0.27 1/308 5/77 0/308
11.0.26 6/263 7/105 2/263
11.0.25 4/243 3/81 0/243
11.0.24 1/243 1/81 0/243
11.0.23 8/567 4/81 2/567
11.0.22 10/951 3/138 1/951
The crashes for 11.0.29 happen in:
- org.apache.tomcat.util.net.TestClientCert FAILED (crashed)
1 nio2 adopt_jdk25-rhel10.x86_64-tcnative-2.0.17-410-1
- org.apache.tomcat.util.net.TestSSLHostConfigCompat FAILED (crashed)
1 nio2 adopt_jdk21-sles15.x86_64-tcnative-2.0.17-400-1
- org.apache.tomcat.util.net.TestSsl FAILED (crashed)
1 nio2 oracle_jdk27-rhel8.x86_64-tcnative-2.0.17-400-1
- org.apache.tomcat.util.net.ocsp.TestOcspEnabled FAILED (crashed)
1 nio2 rh_jdk21-rhel10.x86_64-tcnative-2.0.17-400-1
The only non-crash failure:
- org.apache.tomcat.util.net.ocsp.TestOcspSoftFailTryLater FAILED
1 nio2 adopt_jdk25-rhel8.x86_64-jsse
Native stacks for example:
#0 0x00007fa1a88d4a74 __pthread_kill_implementation
(libc.so.6 + 0x9aa74)
#1 0x00007fa1a887b37e raise (libc.so.6 + 0x4137e)
#2 0x00007fa1a8862945 abort (libc.so.6 + 0x28945)
#3 0x00007fa1a88639b9 __libc_message_impl.cold
(libc.so.6 + 0x299b9)
#4 0x00007fa1a88dec25 malloc_printerr (libc.so.6 +
0xa4c25)
#5 0x00007fa1a88e0e20 _int_free_merge_chunk (libc.so.6
+ 0xa6e20)
#6 0x00007fa1a88e1142 _int_free (libc.so.6 + 0xa7142)
#7 0x00007fa1a88e39ce free (libc.so.6 + 0xa99ce)
#8 0x00007fa1884845f3 apr_allocator_destroy
(libapr-1-KPDT.so.0 + 0x235f3)
#9 0x00007fa188484db0 apr_pool_terminate
(libapr-1-KPDT.so.0 + 0x23db0)
...
#0 0x00007f595d58fa74 __pthread_kill_implementation
(libc.so.6 + 0x9aa74)
#1 0x00007f595d53637e raise (libc.so.6 + 0x4137e)
#2 0x00007f595d51d945 abort (libc.so.6 + 0x28945)
#3 0x00007f595d51e9b9 __libc_message_impl.cold
(libc.so.6 + 0x299b9)
#4 0x00007f595d599c25 malloc_printerr (libc.so.6 +
0xa4c25)
#5 0x00007f595d59be20 _int_free_merge_chunk (libc.so.6
+ 0xa6e20)
#6 0x00007f595d59c142 _int_free (libc.so.6 + 0xa7142)
#7 0x00007f595d59e9ce free (libc.so.6 + 0xa99ce)
#8 0x00007f593c09f5f3 apr_allocator_destroy
(libapr-1-KPDT.so.0 + 0x235f3)
#9 0x00007f593c09fdb0 apr_pool_terminate
(libapr-1-KPDT.so.0 + 0x23db0)
I do not know the Java frames.
The crashes look roughly like those I observe (sporadically) for years.
I vaguely remember from looking at gdb details, that pointers in the
pool or allocator were no longer valid, which looked like threads using
the structures after other threads had already destroyed/terminated
them. All documented in old mails, nothing secret here.
Best regards,
Rainer
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
