Been using HTTPAPI for years, thank you!
We are currently on version 1.39.
Every so often we receive the following error when posting to endpoints, and we are trying to track down the root of the error.
gsk_env_init: (GSKit) Storage allocation request failed.
Some of our jobs are continually running, and we're wondering if something didn't close down properly when ssl connection issues arise.
When our error occurs, we've noticed we've also received prior errors in the job:
SSL Handshake: (GSKit) Peer not recognized or badly formatted message received.
SSL Handshake: (GSKit) I/O: A connection with a remote socket was reset by that
Has anyone experienced anything similar?
We're looking at some of the source code "COMMSSLR4", and we noticed we only issue a "gsk_secure_soc_close" after a timeout from a "gsk_secure_soc_init".
We are wondering if that it always should be issued if there were any errors during a "gsk_secure_soc_init"? Our thought here is that we may have some half baked work still in memory.
Code: Select all
0410.00 c eval rc = gsk_secure_soc_init(wwSslh)
0411.00 c if rc <> GSK_OK
0412.00 c if rc = GSK_IBMI_ERROR_TIMED_OUT
0413.00 c callp SetError(HTTP_SSTIMO: 'Time out during '
0414.00 c 'SSL handshake')
0415.00 c callp gsk_secure_soc_close(wwsslh)
0416.00 c return *OFF
0417.00 c endif
0418.00 c callp SetError(HTTP_SSSNFD: 'SSL Handshake: '
0419.00 c ssl_error(rc))
0420.00 c return *OFF
0421.00 c endif