[2024.2.0.0-b116][AlmaLinux9][Connection Pooling][YBM] Connections erroring out during query execution with error 'SSL SYSCALL error: EOF detected'. #24979
Assignees
Labels
Comments
yugabyte-ci
added
jira-originated
kind/bug
manav-yb
pushed a commit
that referenced
this issue
Nov 20, 2024
…tion manager Summary: Set pthread_attr_setstacksize to 512 KB in ysql connection manager. This is to fix crashes involving Alma 9 machines. ``` #0 0x000055e430191fd7 in tcmalloc::tcmalloc_internal::PageTracker::Get(tcmalloc::tcmalloc_internal::Length) () #1 0x000055e430192774 in tcmalloc::tcmalloc_internal::HugePageFiller<tcmalloc::tcmalloc_internal::PageTracker>::TryGet(tcmalloc::tcmalloc_internal::Length, unsigned long) () #2 0x000055e430160b3a in tcmalloc::tcmalloc_internal::HugePageAwareAllocator::New(tcmalloc::tcmalloc_internal::Length, unsigned long) () #3 0x000055e4301437a2 in void* tcmalloc::tcmalloc_internal::SampleifyAllocation<tcmalloc::tcmalloc_internal::Static, tcmalloc::tcmalloc_internal::TCMallocPolicy<tcmalloc::tcmalloc_internal::MallocOomPolicy, tcmalloc::tcmalloc_internal::MallocAlignPolicy, tcmalloc::tcmalloc_internal::AllocationAccessHotPolicy, tcmalloc::tcmalloc_internal::InvokeHooksPolicy, tcmalloc::tcmalloc_internal::LocalNumaPartitionPolicy> >(tcmalloc::tcmalloc_internal::Static&, tcmalloc::tcmalloc_internal::TCMallocPolicy<tcmalloc::tcmalloc_internal::MallocOomPolicy, tcmalloc::tcmalloc_internal::MallocAlignPolicy, tcmalloc::tcmalloc_internal::AllocationAccessHotPolicy, tcmalloc::tcmalloc_internal::InvokeHooksPolicy, tcmalloc::tcmalloc_internal::LocalNumaPartitionPolicy>, unsigned long, unsigned long, unsigned long, void*, tcmalloc::tcmalloc_internal::Span*, unsigned long*) () #4 0x000055e43014339b in void* slow_alloc<tcmalloc::tcmalloc_internal::TCMallocPolicy<tcmalloc::tcmalloc_internal::MallocOomPolicy, tcmalloc::tcmalloc_internal::MallocAlignPolicy, tcmalloc::tcmalloc_internal::AllocationAccessHotPolicy, tcmalloc::tcmalloc_internal::InvokeHooksPolicy, tcmalloc::tcmalloc_internal::LocalNumaPartitionPolicy>, decltype(nullptr)>(tcmalloc::tcmalloc_internal::TCMallocPolicy<tcmalloc::tcmalloc_internal::MallocOomPolicy, tcmalloc::tcmalloc_internal::MallocAlignPolicy, tcmalloc::tcmalloc_internal::AllocationAccessHotPolicy, tcmalloc::tcmalloc_internal::InvokeHooksPolicy, tcmalloc::tcmalloc_internal::LocalNumaPartitionPolicy>, unsigned long, decltype(nullptr)) () #5 0x000055e4301404e6 in malloc () #6 0x00007fc3a67b1d7e in ssl3_setup_write_buffer () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #7 0x00007fc3a67ae824 in do_ssl3_write () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #8 0x00007fc3a67ae3e1 in ssl3_write_bytes () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #9 0x00007fc3a67cc251 in ssl3_do_write () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #10 0x00007fc3a67c1a6a in state_machine () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #11 0x000055e43013d303 in mm_tls_handshake_cb (handle=<optimized out>) at ../../src/odyssey/third_party/machinarium/sources/tls.c:453 #12 0x000055e43013b9e7 in mm_epoll_step (poll=0x37beffd71a60, timeout=<optimized out>) at ../../src/odyssey/third_party/machinarium/sources/epoll.c:79 #13 0x000055e43013b386 in mm_loop_step (loop=0x37beffd72980) at ../../src/odyssey/third_party/machinarium/sources/loop.c:64 #14 machine_main (arg=0x37beffd72780) at ../../src/odyssey/third_party/machinarium/sources/machine.c:56 #15 0x00007fc3a5e89c02 in start_thread (arg=<optimized out>) at pthread_create.c:443 #16 0x00007fc3a5f0ec40 in clone3 () at ../sysdeps/unix/sysv/linux/x86_64/clone3.S:81``` Note that, this 512 KB value is the same as the value used in tserver and master processes via the min_thread_stack_size_bytes GFlag introduced in https://phorge.dev.yugabyte.com/D38053. Jira: DB-13388 Test Plan: Jenkins: enable connection manager, all tests Reviewers: skumar, stiwary Reviewed By: stiwary Subscribers: yql Differential Revision: https://phorge.dev.yugabyte.com/D40087
manav-yb
pushed a commit
that referenced
this issue
Nov 20, 2024
…KB in ysql connection manager Summary: Original commit: None / D40087 Set pthread_attr_setstacksize to 512 KB in ysql connection manager. This is to fix crashes involving Alma 9 machines. ``` #0 0x000055e430191fd7 in tcmalloc::tcmalloc_internal::PageTracker::Get(tcmalloc::tcmalloc_internal::Length) () #1 0x000055e430192774 in tcmalloc::tcmalloc_internal::HugePageFiller<tcmalloc::tcmalloc_internal::PageTracker>::TryGet(tcmalloc::tcmalloc_internal::Length, unsigned long) () #2 0x000055e430160b3a in tcmalloc::tcmalloc_internal::HugePageAwareAllocator::New(tcmalloc::tcmalloc_internal::Length, unsigned long) () #3 0x000055e4301437a2 in void* tcmalloc::tcmalloc_internal::SampleifyAllocation<tcmalloc::tcmalloc_internal::Static, tcmalloc::tcmalloc_internal::TCMallocPolicy<tcmalloc::tcmalloc_internal::MallocOomPolicy, tcmalloc::tcmalloc_internal::MallocAlignPolicy, tcmalloc::tcmalloc_internal::AllocationAccessHotPolicy, tcmalloc::tcmalloc_internal::InvokeHooksPolicy, tcmalloc::tcmalloc_internal::LocalNumaPartitionPolicy> >(tcmalloc::tcmalloc_internal::Static&, tcmalloc::tcmalloc_internal::TCMallocPolicy<tcmalloc::tcmalloc_internal::MallocOomPolicy, tcmalloc::tcmalloc_internal::MallocAlignPolicy, tcmalloc::tcmalloc_internal::AllocationAccessHotPolicy, tcmalloc::tcmalloc_internal::InvokeHooksPolicy, tcmalloc::tcmalloc_internal::LocalNumaPartitionPolicy>, unsigned long, unsigned long, unsigned long, void*, tcmalloc::tcmalloc_internal::Span*, unsigned long*) () #4 0x000055e43014339b in void* slow_alloc<tcmalloc::tcmalloc_internal::TCMallocPolicy<tcmalloc::tcmalloc_internal::MallocOomPolicy, tcmalloc::tcmalloc_internal::MallocAlignPolicy, tcmalloc::tcmalloc_internal::AllocationAccessHotPolicy, tcmalloc::tcmalloc_internal::InvokeHooksPolicy, tcmalloc::tcmalloc_internal::LocalNumaPartitionPolicy>, decltype(nullptr)>(tcmalloc::tcmalloc_internal::TCMallocPolicy<tcmalloc::tcmalloc_internal::MallocOomPolicy, tcmalloc::tcmalloc_internal::MallocAlignPolicy, tcmalloc::tcmalloc_internal::AllocationAccessHotPolicy, tcmalloc::tcmalloc_internal::InvokeHooksPolicy, tcmalloc::tcmalloc_internal::LocalNumaPartitionPolicy>, unsigned long, decltype(nullptr)) () #5 0x000055e4301404e6 in malloc () #6 0x00007fc3a67b1d7e in ssl3_setup_write_buffer () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #7 0x00007fc3a67ae824 in do_ssl3_write () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #8 0x00007fc3a67ae3e1 in ssl3_write_bytes () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #9 0x00007fc3a67cc251 in ssl3_do_write () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #10 0x00007fc3a67c1a6a in state_machine () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #11 0x000055e43013d303 in mm_tls_handshake_cb (handle=<optimized out>) at ../../src/odyssey/third_party/machinarium/sources/tls.c:453 #12 0x000055e43013b9e7 in mm_epoll_step (poll=0x37beffd71a60, timeout=<optimized out>) at ../../src/odyssey/third_party/machinarium/sources/epoll.c:79 #13 0x000055e43013b386 in mm_loop_step (loop=0x37beffd72980) at ../../src/odyssey/third_party/machinarium/sources/loop.c:64 #14 machine_main (arg=0x37beffd72780) at ../../src/odyssey/third_party/machinarium/sources/machine.c:56 #15 0x00007fc3a5e89c02 in start_thread (arg=<optimized out>) at pthread_create.c:443 #16 0x00007fc3a5f0ec40 in clone3 () at ../sysdeps/unix/sysv/linux/x86_64/clone3.S:81``` Note that, this 512 KB value is the same as the value used in tserver and master processes via the min_thread_stack_size_bytes GFlag introduced in https://phorge.dev.yugabyte.com/D38053. Jira: DB-13388 Test Plan: Jenkins: enable connection manager, all tests Reviewers: skumar, stiwary Reviewed By: stiwary Subscribers: yql Differential Revision: https://phorge.dev.yugabyte.com/D40088
manav-yb
pushed a commit
that referenced
this issue
Nov 20, 2024
…2 KB in ysql connection manager Summary: Original commit: None / D40088 Set pthread_attr_setstacksize to 512 KB in ysql connection manager. This is to fix crashes involving Alma 9 machines. ``` #0 0x000055e430191fd7 in tcmalloc::tcmalloc_internal::PageTracker::Get(tcmalloc::tcmalloc_internal::Length) () #1 0x000055e430192774 in tcmalloc::tcmalloc_internal::HugePageFiller<tcmalloc::tcmalloc_internal::PageTracker>::TryGet(tcmalloc::tcmalloc_internal::Length, unsigned long) () #2 0x000055e430160b3a in tcmalloc::tcmalloc_internal::HugePageAwareAllocator::New(tcmalloc::tcmalloc_internal::Length, unsigned long) () #3 0x000055e4301437a2 in void* tcmalloc::tcmalloc_internal::SampleifyAllocation<tcmalloc::tcmalloc_internal::Static, tcmalloc::tcmalloc_internal::TCMallocPolicy<tcmalloc::tcmalloc_internal::MallocOomPolicy, tcmalloc::tcmalloc_internal::MallocAlignPolicy, tcmalloc::tcmalloc_internal::AllocationAccessHotPolicy, tcmalloc::tcmalloc_internal::InvokeHooksPolicy, tcmalloc::tcmalloc_internal::LocalNumaPartitionPolicy> >(tcmalloc::tcmalloc_internal::Static&, tcmalloc::tcmalloc_internal::TCMallocPolicy<tcmalloc::tcmalloc_internal::MallocOomPolicy, tcmalloc::tcmalloc_internal::MallocAlignPolicy, tcmalloc::tcmalloc_internal::AllocationAccessHotPolicy, tcmalloc::tcmalloc_internal::InvokeHooksPolicy, tcmalloc::tcmalloc_internal::LocalNumaPartitionPolicy>, unsigned long, unsigned long, unsigned long, void*, tcmalloc::tcmalloc_internal::Span*, unsigned long*) () #4 0x000055e43014339b in void* slow_alloc<tcmalloc::tcmalloc_internal::TCMallocPolicy<tcmalloc::tcmalloc_internal::MallocOomPolicy, tcmalloc::tcmalloc_internal::MallocAlignPolicy, tcmalloc::tcmalloc_internal::AllocationAccessHotPolicy, tcmalloc::tcmalloc_internal::InvokeHooksPolicy, tcmalloc::tcmalloc_internal::LocalNumaPartitionPolicy>, decltype(nullptr)>(tcmalloc::tcmalloc_internal::TCMallocPolicy<tcmalloc::tcmalloc_internal::MallocOomPolicy, tcmalloc::tcmalloc_internal::MallocAlignPolicy, tcmalloc::tcmalloc_internal::AllocationAccessHotPolicy, tcmalloc::tcmalloc_internal::InvokeHooksPolicy, tcmalloc::tcmalloc_internal::LocalNumaPartitionPolicy>, unsigned long, decltype(nullptr)) () #5 0x000055e4301404e6 in malloc () #6 0x00007fc3a67b1d7e in ssl3_setup_write_buffer () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #7 0x00007fc3a67ae824 in do_ssl3_write () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #8 0x00007fc3a67ae3e1 in ssl3_write_bytes () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #9 0x00007fc3a67cc251 in ssl3_do_write () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #10 0x00007fc3a67c1a6a in state_machine () from /home/centos/code/local_testing/conn_manager_ssl_auth/yugabyte-b124/bin/../lib/yb-thirdparty/libssl.so.3 #11 0x000055e43013d303 in mm_tls_handshake_cb (handle=<optimized out>) at ../../src/odyssey/third_party/machinarium/sources/tls.c:453 #12 0x000055e43013b9e7 in mm_epoll_step (poll=0x37beffd71a60, timeout=<optimized out>) at ../../src/odyssey/third_party/machinarium/sources/epoll.c:79 #13 0x000055e43013b386 in mm_loop_step (loop=0x37beffd72980) at ../../src/odyssey/third_party/machinarium/sources/loop.c:64 #14 machine_main (arg=0x37beffd72780) at ../../src/odyssey/third_party/machinarium/sources/machine.c:56 #15 0x00007fc3a5e89c02 in start_thread (arg=<optimized out>) at pthread_create.c:443 #16 0x00007fc3a5f0ec40 in clone3 () at ../sysdeps/unix/sysv/linux/x86_64/clone3.S:81``` Note that, this 512 KB value is the same as the value used in tserver and master processes via the min_thread_stack_size_bytes GFlag introduced in https://phorge.dev.yugabyte.com/D38053. Jira: DB-13388 Test Plan: Jenkins: enable connection manager, all tests Reviewers: skumar, stiwary Reviewed By: stiwary Subscribers: yql Differential Revision: https://phorge.dev.yugabyte.com/D40089
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Jira Link: DB-13388
The text was updated successfully, but these errors were encountered: