sinknesha opened a new issue, #1808: URL: https://github.com/apache/cloudberry/issues/1808
Hello, We are frequently encountering the message "interconnect may have encountered a network error, please check your network" in the segment logs. Additionally, we have observed segment failures on several occasions, with the same error reported in the logs. We would appreciate your input and suggestions on how to resolve this issue. Aside from the interconnect error, we have been experiencing segmentation fault (signal 11 SIGSEGV) related crashes, mainly with CTE queries and more recently with the update statement. Steps taken for optimization so far interconnect warnings/errors: · Server health and network health have been checked, no noticeable issues have been reported. · The gp_max_packet_size value has been reduced to 1420 from the default value of 8192, to align with the VM’s MTU of 1500. However, this adjustment does not seem to have helped much and requires further data, along with addressing the interconnect error. Our current cluster settings: gp_max_packet_size = 8192 gp_interconnect_type = udpifc gp_interconnect_transmit_timeout = 1hr gp_fts_probe_timeout = 40 seconds gp_interconnect_type = udpifc ## system net.core.rmem_max = 4194304 net.core.rmem_default = 212992 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
