bad resource id msg from portable cas
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
EPICS Base |
Won't Fix
|
Low
|
mdavidsaver |
Bug Description
Fro Kay:
For what it's worth, we're seeing these messages on a CA Gateway at the SNS:
-------------
CAS Request: ... on ....: cmd=2 cid=1 typ=20 cnt=1 psz=0 avail=130
CAS:
Mar 04 08:20:39 !!! Errlog message received (message is above)
bad resource id in "../../
1988
Mar 04 08:20:39 !!! Errlog message received (message is above)
filename=
Bad resource identifier - unexpected problem with client's input - forcing
disconnect
-------------
The messages appear to be caused by pure Java (CAJ) clients running on
Windows, Linux and Mac OS X.
The gateway is version 2.0.4.0, based on EPICS R3.14.11.
CAJ is 1.1.7 with JCA 2.3.3 to 2.3.5 depending on the specific client, can't
say for sure which one is which.
We are not sure it's a severe problem. The gateway continues to run.
Since we don't use CAJ on the accelerator network, only on the office
network where all access is through a read-only gateway, we can't spend much
time on debugging this issue, but sites that depend on CAJ for a production
setup might want to check if they get similar messages.
tags: | added: cas gateway portable |
Changed in epics-base: | |
status: | New → Incomplete |
importance: | Undecided → Low |
assignee: | nobody → mdavidsaver (mdavidsaver) |
From Steven Hartman:
It appears that these "bad resource" messages correspond with a
big increase in network traffic on the public interface of the
gateway machine: from 3,000 packets/sec to 4,000,000 pk/s receive
and 8,000 pk/s to 500,000 pk/s transmit. (This gateway provides
read-only access to clients outside the control system firewall.
The network interface showing the increase is the one on the
"office" side, not those on the control system side which seem fine.)
This is gateway-2-0-4-0 on EPICS 3.14.11 on RHEL 4-8, x86 32-bit.