Closed longlong10086 closed 2 years ago
I noticed that in the email you send to the security list, there was an error
msg="Error making datagram connection to Job Manager" reason="No such file or directory" path="/vol6/home/globus/softs/gct-6.2/var/lib/globus/gram_job_state/globus/test/fork.28be48cb.sock"
and line155 in globus-job-manager-script.pl
I'm not an expert on GRAM but perhaps the directory where that socket is expected doesn't exist or isn't writable for the job manager?
Thanks for your reply.
Sometimes Ican find the file "/vol6/home/globus/softs/gct-6.2/var/lib/globus/gram_job_state/globus/test/fork.28be48cb.sock" ,sometimes can't , and now log changes .
event=gram.send_job.end level=WARN status=-3 errno=111 msg="Error making datagram connection to Job Manager" reason="Connectio n refused" path="/vol6/home/globus/softs/gct-6.2/var/lib/globus/gram_job_state/globus/nscctj-1/fork.28be48cb.sock"
------------------ 原始邮件 ------------------ 发件人: "gridcf/gct" @.>; 发送时间: 2021年10月27日(星期三) 下午4:06 @.>; @.**@.>; 主题: Re: [gridcf/gct] GRAM Job failed because the job manager failed to open stdout (Issue #172)
I noticed that in the email you send to the security list, there was an error msg="Error making datagram connection to Job Manager" reason="No such file or directory" path="/vol6/home/globus/softs/gct-6.2/var/lib/globus/gram_job_state/globus/test/fork.28be48cb.sock" and line155 in globus-job-manager-script.pl I'm not an expert on GRAM but perhaps the directory where that socket is expected doesn't exist or isn't writable for the job manager?
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
Hi all, the matter in question might be understood from this Wiki page:
https://wiki.egi.eu/wiki/Tools/Manuals/TS77
As that documentation was for older versions of Globus, it may not explain such issues with recent versions, though.
I deployed a cluster with gct-6.2.
No matter on the slave node or the master node, I met the error "GRAM Job failed because the job manager failed to open stdout (error code 73) Details:", when executing the command " globus-job-run slave-node/jobmanager-fork-poll -np 1 /bin/hostnamee" .
Now I try parameter "-stdout /tmp/1.log", likely "globus-job-run slave-node/jobmanager-fork-poll -np 1 -stdout /tmp/1.log /bin/hostnamee" , I get right result in /tmp/1.log , but the error above still appears in the terminal .
How to fix ti ?