Experimental Physics and
| |||||||||||||||
|
Hi all, Our telescope control system is on an MVME2700, EPICS 3.14.12.7 and RTEMS 4.10.2. We've seen a few events in the last several months of starved mbuf's. There are two main failure paths: 1) If we burden the IOC by calling debug utilities with higher interest levels like dbcar "" 10, or casr 2, we can see CA Client screens go white. Sometimes the system will recover by itself over time. 2) Without iocsh or human interactions, the system can start screaming about mbufs before halting without a stack trace. We nfsmount a single directory in the startup script. These RPCIO messages below are related to the nfs server on RTEMS I believe. Should we be resizing the default mbuf capacity? If so, how can we best do this? Thanks, -Matt Nov 30 16:28:40 : RPCIO: server '10.2.2.57' not responding - still trying Nov 30 16:28:43 : NFS (proc 1) - RPC: Timed out Nov 30 16:28:47 : Still waiting for mbuf cluster. Nov 30 16:28:47 : Still waiting for mbuf cluster. Nov 30 16:28:47 : Still waiting for mbuf cluster. Nov 30 16:28:47 : Still waiting for mbuf cluster. Nov 30 16:28:47 : Still waiting for mbuf cluster. Nov 30 16:28:47 : Still waiting for mbuf cluster. Nov 30 16:28:47 : Still waiting for mbuf cluster. Nov 30 16:28:47 : RPCIO: server '10.2.2.57' not responding - still trying Nov 30 16:28:47 : Still waiting for mbuf cluster. Nov 30 16:28:48 : Still waiting for mbuf cluster. Nov 30 16:28:48 : Still waiting for mbuf cluster. Nov 30 16:28:48 : Still waiting for mbuf cluster. Nov 30 16:28:51 : NFS (proc 1) - RPC: Timed out Nov 30 16:28:54 : RPCIO: server '10.2.2.57' not responding - still trying
| ||||||||||||||
ANJ, 04 Dec 2018 |
·
Home
·
News
·
About
·
Base
·
Modules
·
Extensions
·
Distributions
·
Download
·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing · |