Experimental Physics and Industrial Control System
|
Hi Torsten,
The default timeout for the catools to connect is 1 second. Increasing that (using
-w 5 for example) would have let
cainfo send out more search packets, thus have more chances to connect if they were being discarded at random. I can’t explain why
caget would consistently behave any differently than
cainfo though. Both programs give up when trying to connect to a nonexistent PV after between 1.013 - 1.035 seconds for me on macOS.
- Andrew
(Just to avoid confusion:
netstat was run on the Linux box,
pvget on the Mac)
)
On 11/09/20 10:44, Torsten Bögershausen via Core-talk wrote:
Hej Timo and all,
(Starting top-posting)
pvget gives me this:
pvget -d IOC:m1
2020-09-11T10:35:47.820 Creating datagram socket from: 0.0.0.0:60988.
2020-09-11T10:35:47.821 Broadcast address #0: 192.168.xxx.255:5076. (not unicast)
2020-09-11T10:35:47.821 Setting up UDP for interface 192.168.xxx.79/255.255.255.0, broadcast 192.168.xxx.255, dest <none>.
2020-09-11T10:35:47.821 Creating datagram socket from: 192.168.xxx.79:5076.
2020-09-11T10:35:47.821 Creating datagram socket from: 192.168.xxx.255:5076.
2020-09-11T10:35:47.821 Creating datagram socket from: 224.0.0.128:5076.
2020-09-11T10:35:47.822 Local multicast enabled on 127.0.0.1/224.0.0.128:5076.
2020-09-11T10:35:47.822 Sending 52 bytes 0.0.0.0:60988 -> 192.168.xxx.255:5076.
2020-09-11T10:35:47.822 UDP Client Rx (52) 192.168.xxx.255:5076 <- 192.168.xxx.79:60988
Waiting...
2020-09-11T10:35:48.068 Sending 52 bytes 0.0.0.0:60988 -> 192.168.xxx.255:5076.
2020-09-11T10:35:48.068 UDP Client Rx (52) 192.168.xxx.255:5076 <- 192.168.xxx.79:60988
2020-09-11T10:35:48.555 Sending 52 bytes 0.0.0.0:60988 -> 192.168.xxx.255:5076.
2020-09-11T10:35:48.556 UDP Client Rx (52) 192.168.xxx.255:5076 <- 192.168.xxx.79:60988
2020-09-11T10:35:49.528 Sending 52 bytes 0.0.0.0:60988 -> 192.168.xxx.255:5076.
2020-09-11T10:35:49.528 UDP Client Rx (52) 192.168.xxx.255:5076 <- 192.168.xxx.79:60988
2020-09-11T10:35:51.473 Sending 52 bytes 0.0.0.0:60988 -> 192.168.xxx.255:5076.
2020-09-11T10:35:51.473 UDP Client Rx (52) 192.168.xxx.255:5076 <- 192.168.xxx.79:60988
Timeout
IOC:m1 2020-09-11T10:35:52.824 UDP socket 0.0.0.0:0 closed.
2020-09-11T10:35:52.824 UDP socket 192.168.xxx.79:5076 closed.
2020-09-11T10:35:52.824 UDP socket 192.168.xxx.255:5076 closed.
2020-09-11T10:35:52.825 UDP socket 224.0.0.128:5076 closed.
And this is from netstat, leaving out the "ESTABLISHED" sockets:
netstat -a -n -t | grep 506
tcp 0 0 0.0.0.0:5064 0.0.0.0:* LISTEN
netstat -a -n -u | grep 506
udp 0 0 0.0.0.0:5064 0.0.0.0:*
udp 0 0 0.0.0.0:5065 0.0.0.0:*
On 11/09/20 10:19, Timo Korhonen wrote:
Torsten, do you have pvget on your Mac? If yes, trying
pvget -d IOC:m1
could give you some hints on what is going on the network. (-d for debug).
Assuming your IOC is PVA-capable, that is.
Just my 2 cents __
Timo
On 11/09/20 10:12, "Core-talk on behalf of Torsten Bögershausen via Core-talk" <core-talk-bounces at aps.anl.gov on behalf of
core-talk at aps.anl.gov> wrote:
On 11/09/20 09:09, Torsten Bögershausen via Core-talk wrote:
>
>
> On 11/09/20 08:53, Ralph Lange via Core-talk wrote:
>> On Fri, 11 Sep 2020 at 07:40, Torsten Bögershausen via Core-talk
>> <core-talk at aps.anl.gov <mailto:core-talk at aps.anl.gov>> wrote:
>>
>> Trying to find the box that is running the IOC ddidn't work:
>>
>> ~/shared_EEE/projects/epics/MCAG_setupMotionDemo.200109-base-7.0.3 $
>> cainfo IOC:m1
>> Channel connect timed out: 'IOC:m1' not found.
>>
>> But caget does work:
>>
>> ~/shared_EEE/projects/epics/MCAG_setupMotionDemo.200109-base-7.0.3 $
>> caget IOC:m1
>> IOC:m1 6.01
>>
>>
>> This is under MacOS, EPICS base 7.0.3.
>> env | grep EPICS shows nothing special.
>>
>>
>> Unusual indeed.
>>
>> Please try 'which' on the commands.
>> Maybe one of them is a 'convenience' script that sets the environment
>> before running the binary from Base.
>>
>> Cheers,
>> ~Ralph
>>
>
> Side note:
> The path is obfuscated,
> (and I will reboot the machine, it's Friday)
>
> ~/somePath $ which caget
> /Users/torstenbogershausen/somePath/epics/base/bin/darwin-x86/caget
> ~/somePath $ type caget
> caget is hashed
> (/Users/torstenbogershausen/somePath/epics/base/bin/darwin-x86/caget)
> ~/somePath $ type cainfo
> cainfo is hashed
> (/Users/torstenbogershausen/somePath/epics/base/bin/darwin-x86/cainfo)
> ~/somePath $ which cainfo
> /Users/torstenbogershausen/somePath/epics/base/bin/darwin-x86/cainfo
> ~/shared_EEE/somePath $
Thanks Ralph,
And here some more useless information...
- Rebooted the Mac
- Found the IOC on a linux machine,
being once started in the background as part of an automated test.
Most probably we have settings like this:
EPICS_CA_ADDR_LIST=127.0.0.1
EPICS_CA_AUTO_ADDR_LIST=NO
export EPICS_CA_ADDR_LIST EPICS_CA_AUTO_ADDR_LIST
Then we have a shell script to start the IOC, running
Base R7.0.4.1
The Linux box is connected to WLAN.
On the (rebooted) Mac (WLAN as well) the following
happens:
EPICS Base 7.0.3 (on the Mac):
caget finds the PV sometimes, roughly 2-3 times of 10 tries
cainfo does not find it.
EPICS Base R3.15.5:
caget fails always, and so does cainfo
I don't know if we should put more time on this issue ?
--
Complexity comes for free, simplicity you have to work for.
|
- References:
- caget working, but cainfo not ? Torsten Bögershausen via Core-talk
- Re: caget working, but cainfo not ? Ralph Lange via Core-talk
- Re: caget working, but cainfo not ? Torsten Bögershausen via Core-talk
- Re: caget working, but cainfo not ? Torsten Bögershausen via Core-talk
- Re: caget working, but cainfo not ? Timo Korhonen via Core-talk
- Re: caget working, but cainfo not ? Torsten Bögershausen via Core-talk
- Re: caget working, but cainfo not ? Torsten Bögershausen via Core-talk
- Navigate by Date:
- Prev:
Re: caget working, but cainfo not ? Torsten Bögershausen via Core-talk
- Next:
Compiler warning question Mark Rivers via Core-talk
- Index:
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
<2020>
2021
2022
2023
2024
- Navigate by Thread:
- Prev:
Re: caget working, but cainfo not ? Torsten Bögershausen via Core-talk
- Next:
Compiler warning question Mark Rivers via Core-talk
- Index:
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
<2020>
2021
2022
2023
2024
|
ANJ, 15 Sep 2020 |
·
Home
·
News
·
About
·
Base
·
Modules
·
Extensions
·
Distributions
·
Download
·
·
Search
·
EPICS V4
·
IRMIS
·
Talk
·
Bugs
·
Documents
·
Links
·
Licensing
·
|