hey guys...
got a much simpler issue/question...
I thought I had been running the remote cmd within the screen named
session.. upon further testing.. I may not have actually been running
it in the screen session.. perhaps outside of the session..
I think this works
//--fire up the named scr
On 11/14/2016 12:01 PM, Tom Horsley wrote:
ssh goes to fantastic amounts of trouble to invent
ridiculous file descriptors it can wait on till
they close
No it doesn't. ssh has a simple, well-defined behavior. It doesn't
create non-standard (read "ridiculous") file descriptors. If you're
ha
On Mon, 14 Nov 2016 14:13:48 -0500
bruce wrote:
> It only detaches after the process completes!!
ssh goes to fantastic amounts of trouble to invent
ridiculous file descriptors it can wait on till
they close (specifically to prevent you from
exiting an ssh session and leaving anything running).
I
On 11/14/2016 11:13 AM, bruce wrote:
> Hey Rick...
>
> Morning dude! Thanks for the reply.
>
> I've tried a bunch of different combinations, including the one you
> posted.. What seems to be happening, is that I can get it to run. The
> tmp file is generated with the data but the session doesn't
Hey Rick...
Morning dude! Thanks for the reply.
I've tried a bunch of different combinations, including the one you
posted.. What seems to be happening, is that I can get it to run. The
tmp file is generated with the data but the session doesn't
immeadiately detach. It only detaches after the pro
On 11/14/2016 03:01 AM, bruce wrote:
> Hi Jakub
>
> So you're saying
>
ssh crawl_user@192.81.214.49 'screen -r "testname" -X "`ls -al
/crawl_tmp/* > aa.tmp`" '
> is not the correct approach to running a remote cmd in a screen
> session, and that the cmd wrapped in the backticks is
Hi Jakub
So you're saying
>>> ssh crawl_user@192.81.214.49 'screen -r "testname" -X "`ls -al
>>> /crawl_tmp/* > aa.tmp`" '
is not the correct approach to running a remote cmd in a screen
session, and that the cmd wrapped in the backticks is run on the local
machine, not the remote machine???
On 11/14/2016 02:17 AM, bruce wrote:
I think I'm getting closer to an actual soln...
** Notice the backticks around the remote cmd to be run
-without the ticks, the cmd doesn't appear to work
ssh crawl_user@192.81.214.49 'screen -r "testname" -X "`nohup ls -al
/crawl_tmp/* > aa.tmp`" '
ss
I think I'm getting closer to an actual soln...
** Notice the backticks around the remote cmd to be run
-without the ticks, the cmd doesn't appear to work
ssh crawl_user@192.81.214.49 'screen -r "testname" -X "`nohup ls -al
/crawl_tmp/* > aa.tmp`" '
ssh crawl_user@192.81.214.49 'screen -r "
On Sun, 2016-11-13 at 10:09 -0500, bruce wrote:
> The goal, to be able to reattach to a remote screen session, and to
> run a cmd in the remote screen session, and have the cmd return to the
> calling term
Maybe take a look at expect (dnf info expect).
poc
___
Followup.
So far lots of tests/no success:
To refresh:
The goal, to be able to reattach to a remote screen session, and to
run a cmd in the remote screen session, and have the cmd return to the
calling term
Research/testing I should be able to create a remote screen session,
and then reattach t
Hey Ladies/Gents.
This follows the other thread i posted over the last few weeks on
devOPS stuff for managing groups of remote droplets/vms.
My approach for now to implement clusterSSH as well as screen to be
able to auto gen the remote droplets, and to be able to ssh into them
to view status, as
12 matches
Mail list logo
