Re: iocLogServer, logstash & procServ

["Johnson, Andrew N. via Tech-talk" <tech-talk at aps.anl.gov>]

Hi,

On Jun 9, 2020, at 10:23 PM, 吴煊 via Tech-talk <tech-talk at aps.anl.gov> wrote:

I quote this message as I wish to know whether the issue of logstash dropping connection has been fixed, or how is it going on now.

Both the recent EPICS 7.0.4 and 3.15.8 releases contain a series of changes that Dirk Zimoch (PSI) developed to mitigate the problems with logging as much as is possible over a TCP connection. Logging using Logstash or any other IOC log-server will benefit from these fixes.

HTH,

- Andrew

-----原始邮件-----
发件人:"Kivel Niko (PSI) via Tech-talk" <tech-talk at aps.anl.gov>
发送时间:2019-10-08 19:12:29 (星期二)
收件人: "Alfio Rizzo" <Alfio.Rizzo at esss.se>, "tech-talk at aps.anl.gov" <tech-talk at aps.anl.gov>
抄送: 
主题: RE: iocLogServer, logstash & procServ

Hi Alfio

it's not an issue with logstash, the issue is in Epics base. To my best knowledge, Dirk is working on it. But, I'm not certain about the 'official' release status.

Right now we operate a fixed version of Epics base in production and I get north of 40k caPutLogs per hour w/o problems.

I use the ELK stack for syslog, caputlog and hardware monitoring through icinga2. The total index rate is about 100 /s and it's stable for weeks now. 

Imho it's not a good idea to find a workaround, something as essential as logging should be properly implemented in base.

best

Niko

--
Paul Scherrer Institute
Dr. Niko Kivel

Forschungsstrasse 111

WBGB/014
5232 Villigen PSI
Switzerland

phone: +41 56 310 2226
email: niko.kivel at psi.ch

-- 

Complexity comes for free, simplicity you have to work for.