Re: [systemd-devel] udev event on usb unpluged

2015-08-26 Thread christophe . jalady
 I am not sure about what you mean here. Do you mean that you can't see
 any remove event with `udevadm monitor` when you unplug the cable?

yes, my script never receive the remove event (_but_ receive the add event).


 Also do you really mean a usb cable without any device connected to
 it on the other end?

No, the cable is connected to another device. So I want the remove event when 
the cable is unplugged from one of the device (my device or the other one).


 Unless you have a very specific to match with, I wonder if it's a good
 idea to unload the module by a remove event,
 ...what if there are two cables/devices using the module?

Well, my setup is really specific (small robotic device), so it safe to 
consider only one cable connected.


Also it seems to be g_mass_storage is just the module for setting up
dummy or emulating an mass storage device. Why would you want it to
depend on a usb cable/device?

This is the bigger picture:
My device is a small robotic device. To send a file on it, I that users just 
copy files like they do with  mass-storage usb-key.
So: On usb-plugged: I load the module k_mass_storage: the other device see a 
new mass-storage device appearing, copy files and then umount/unplugged the 
cable.
When the cable is unplugged, I want to trigger an action on my device 
(processing the files copied).

Christophe



- Mail original -
De: Tom Yan tom.t...@gmail.com
À: christophe jalady christophe.jal...@free.fr
Cc: systemd systemd-devel@lists.freedesktop.org
Envoyé: Mercredi 26 Août 2015 04:37:40
Objet: Re: [systemd-devel] udev event on usb unpluged

Also it seems to be g_mass_storage is just the module for setting up
dummy or emulating an mass storage device. Why would you want it to
depend on a usb cable/device?

On 26 August 2015 at 11:21, Tom Yan tom.t...@gmail.com wrote:
 I am not sure about what you mean here. Do you mean that you can't see
 any remove event with `udevadm monitor` when you unplug the cable?
 Also do you really mean a usb cable without any device connected to
 it on the other end?

 By the way the rule you set up is so broad that the script will
 basically be run on every boot or udevadm trigger.

 Unless you have a very specific to match with, I wonder if it's a good
 idea to unload the module by a remove event, because, for example,
 what if there are two cables/devices using the module?

 On 26 August 2015 at 04:17,  christophe.jal...@free.fr wrote:
 Hi,

 I would like to receive an event (actually run a script) when I 
 plug/unplugged an usb cable.
 I tried to use udev with this simple rule:
 SUBSYSTEM==usb,RUN+=/var/local/usb-changed.sh

 I do receive the add event when I plug the usb cable: just perfect !

 But I never receive the remove event.
 Any idea why ?

 More information:
 I use linux kernel v3.16.
 When the cable is plugged, I load the kernel module g_mass_storage, this 
 work great. My need is to unload the module when the usb cable is unplugged 
 (note it would be even better if I receive an event when the mass-storage is 
 unmounted by the other host, but I do not have any idea where to look for.)

 Thanks by advance for any hint.

 Christophe
 ___
 systemd-devel mailing list
 systemd-devel@lists.freedesktop.org
 http://lists.freedesktop.org/mailman/listinfo/systemd-devel
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel


Re: [systemd-devel] udev event on usb unpluged

2015-08-26 Thread Tom Yan
Can you paste run `udevadm monitor`, plug in your cable/device, wait
for a second, unplug it, and paste the output? Or can you show how you
confirmed your script receive the add event, or how the script
catch the event?

On 26 August 2015 at 15:35,  christophe.jal...@free.fr wrote:
 I am not sure about what you mean here. Do you mean that you can't see
 any remove event with `udevadm monitor` when you unplug the cable?

 yes, my script never receive the remove event (_but_ receive the add 
 event).


 Also do you really mean a usb cable without any device connected to
 it on the other end?

 No, the cable is connected to another device. So I want the remove event 
 when the cable is unplugged from one of the device (my device or the other 
 one).


 Unless you have a very specific to match with, I wonder if it's a good
 idea to unload the module by a remove event,
 ...what if there are two cables/devices using the module?

 Well, my setup is really specific (small robotic device), so it safe to 
 consider only one cable connected.


Also it seems to be g_mass_storage is just the module for setting up
dummy or emulating an mass storage device. Why would you want it to
depend on a usb cable/device?

 This is the bigger picture:
 My device is a small robotic device. To send a file on it, I that users just 
 copy files like they do with  mass-storage usb-key.
 So: On usb-plugged: I load the module k_mass_storage: the other device see a 
 new mass-storage device appearing, copy files and then umount/unplugged the 
 cable.
 When the cable is unplugged, I want to trigger an action on my device 
 (processing the files copied).

 Christophe



 - Mail original -
 De: Tom Yan tom.t...@gmail.com
 À: christophe jalady christophe.jal...@free.fr
 Cc: systemd systemd-devel@lists.freedesktop.org
 Envoyé: Mercredi 26 Août 2015 04:37:40
 Objet: Re: [systemd-devel] udev event on usb unpluged

 Also it seems to be g_mass_storage is just the module for setting up
 dummy or emulating an mass storage device. Why would you want it to
 depend on a usb cable/device?

 On 26 August 2015 at 11:21, Tom Yan tom.t...@gmail.com wrote:
 I am not sure about what you mean here. Do you mean that you can't see
 any remove event with `udevadm monitor` when you unplug the cable?
 Also do you really mean a usb cable without any device connected to
 it on the other end?

 By the way the rule you set up is so broad that the script will
 basically be run on every boot or udevadm trigger.

 Unless you have a very specific to match with, I wonder if it's a good
 idea to unload the module by a remove event, because, for example,
 what if there are two cables/devices using the module?

 On 26 August 2015 at 04:17,  christophe.jal...@free.fr wrote:
 Hi,

 I would like to receive an event (actually run a script) when I 
 plug/unplugged an usb cable.
 I tried to use udev with this simple rule:
 SUBSYSTEM==usb,RUN+=/var/local/usb-changed.sh

 I do receive the add event when I plug the usb cable: just perfect !

 But I never receive the remove event.
 Any idea why ?

 More information:
 I use linux kernel v3.16.
 When the cable is plugged, I load the kernel module g_mass_storage, this 
 work great. My need is to unload the module when the usb cable is unplugged 
 (note it would be even better if I receive an event when the mass-storage 
 is unmounted by the other host, but I do not have any idea where to look 
 for.)

 Thanks by advance for any hint.

 Christophe
 ___
 systemd-devel mailing list
 systemd-devel@lists.freedesktop.org
 http://lists.freedesktop.org/mailman/listinfo/systemd-devel
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel


Re: [systemd-devel] udev event on usb unpluged

2015-08-26 Thread Robert Milasan
On Wed, 26 Aug 2015 14:43:35 +0200 (CEST)
christophe.jal...@free.fr wrote:

 can you show how you confirmed your script receive the add event,
 or how the script catch the event?
 
 Sure: basically I do echo USB_$ACTION  /tmp/jc.log.
 After plug/wait/unplugged/wait I only see lines USB_add.
 See attached my .rules and .sh scripts.
 
 
 Can you paste run `udevadm monitor`, plug in your cable/device, wait
 for a second, unplug it, and paste the output?
 
 Well, this is the trick: I cannot add a connection to my device _and_
 the usb cable connected to it. So I am not able to monitor the
 process directly from my device. Does the scripts provided are enough
 to help you understanding my problem ?
 
 Christophe

They are not referring to your device, but the machine on which you are
running the rule.

In a terminal/console you run: udevadm monitor

The do the removal itself and see what 'udevadm monitor' says.


-- 
Robert Milasan

L3 Support Engineer
SUSE Linux (http://www.suse.com)
email: rmila...@suse.com
GPG fingerprint: B6FE F4A8 0FA3 3040 3402  6FE7 2F64 167C 1909 6D1A
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel


Re: [systemd-devel] udev event on usb unpluged

2015-08-26 Thread christophe . jalady
can you show how you confirmed your script receive the add event, or how the 
script
catch the event?

Sure: basically I do echo USB_$ACTION  /tmp/jc.log.
After plug/wait/unplugged/wait I only see lines USB_add.
See attached my .rules and .sh scripts.


Can you paste run `udevadm monitor`, plug in your cable/device, wait
for a second, unplug it, and paste the output?

Well, this is the trick: I cannot add a connection to my device _and_ the usb 
cable connected to it.
So I am not able to monitor the process directly from my device.
Does the scripts provided are enough to help you understanding my problem ?

Christophe

- Mail original -
De: Tom Yan tom.t...@gmail.com
À: christophe jalady christophe.jal...@free.fr
Cc: systemd systemd-devel@lists.freedesktop.org
Envoyé: Mercredi 26 Août 2015 08:57:12
Objet: Re: [systemd-devel] udev event on usb unpluged

Can you paste run `udevadm monitor`, plug in your cable/device, wait
for a second, unplug it, and paste the output? Or can you show how you
confirmed your script receive the add event, or how the script
catch the event?

On 26 August 2015 at 15:35,  christophe.jal...@free.fr wrote:
 I am not sure about what you mean here. Do you mean that you can't see
 any remove event with `udevadm monitor` when you unplug the cable?

 yes, my script never receive the remove event (_but_ receive the add 
 event).


 Also do you really mean a usb cable without any device connected to
 it on the other end?

 No, the cable is connected to another device. So I want the remove event 
 when the cable is unplugged from one of the device (my device or the other 
 one).


 Unless you have a very specific to match with, I wonder if it's a good
 idea to unload the module by a remove event,
 ...what if there are two cables/devices using the module?

 Well, my setup is really specific (small robotic device), so it safe to 
 consider only one cable connected.


Also it seems to be g_mass_storage is just the module for setting up
dummy or emulating an mass storage device. Why would you want it to
depend on a usb cable/device?

 This is the bigger picture:
 My device is a small robotic device. To send a file on it, I that users just 
 copy files like they do with  mass-storage usb-key.
 So: On usb-plugged: I load the module k_mass_storage: the other device see a 
 new mass-storage device appearing, copy files and then umount/unplugged the 
 cable.
 When the cable is unplugged, I want to trigger an action on my device 
 (processing the files copied).

 Christophe



 - Mail original -
 De: Tom Yan tom.t...@gmail.com
 À: christophe jalady christophe.jal...@free.fr
 Cc: systemd systemd-devel@lists.freedesktop.org
 Envoyé: Mercredi 26 Août 2015 04:37:40
 Objet: Re: [systemd-devel] udev event on usb unpluged

 Also it seems to be g_mass_storage is just the module for setting up
 dummy or emulating an mass storage device. Why would you want it to
 depend on a usb cable/device?

 On 26 August 2015 at 11:21, Tom Yan tom.t...@gmail.com wrote:
 I am not sure about what you mean here. Do you mean that you can't see
 any remove event with `udevadm monitor` when you unplug the cable?
 Also do you really mean a usb cable without any device connected to
 it on the other end?

 By the way the rule you set up is so broad that the script will
 basically be run on every boot or udevadm trigger.

 Unless you have a very specific to match with, I wonder if it's a good
 idea to unload the module by a remove event, because, for example,
 what if there are two cables/devices using the module?

 On 26 August 2015 at 04:17,  christophe.jal...@free.fr wrote:
 Hi,

 I would like to receive an event (actually run a script) when I 
 plug/unplugged an usb cable.
 I tried to use udev with this simple rule:
 SUBSYSTEM==usb,RUN+=/var/local/usb-changed.sh

 I do receive the add event when I plug the usb cable: just perfect !

 But I never receive the remove event.
 Any idea why ?

 More information:
 I use linux kernel v3.16.
 When the cable is plugged, I load the kernel module g_mass_storage, this 
 work great. My need is to unload the module when the usb cable is unplugged 
 (note it would be even better if I receive an event when the mass-storage 
 is unmounted by the other host, but I do not have any idea where to look 
 for.)

 Thanks by advance for any hint.

 Christophe
 ___
 systemd-devel mailing list
 systemd-devel@lists.freedesktop.org
 http://lists.freedesktop.org/mailman/listinfo/systemd-devel

95-usb.rules
Description: Binary data


usb-changed.sh
Description: application/shellscript
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel


Re: [systemd-devel] [ANNOUNCE] systemd-225 around the corner

2015-08-26 Thread David Herrmann
Hi

On Wed, Aug 26, 2015 at 1:04 AM, Thomas H.P. Andersen pho...@gmail.com wrote:
 On Wed, Aug 26, 2015 at 12:40 AM, David Herrmann dh.herrm...@gmail.com 
 wrote:
 Hi

 Trying to continue with our bi-weekly release schedule, we plan to
 release version 225 tomorrow. Please give it a spin and make sure
 there is no major breakage before the release.

 Hi,

 There is a mem leak triggered when running
 src/libsystemd-network/test-dhcp6-client.c. It would be good to know
 if that is from the test code or from dhcp6. I will probably not have
 time to check it myself before the release though.

Both fixed in git now:
e6b18ffaea7d557eec3028a37c043da67a78550c
a00458421dd4b6fcb9b4cdc433ba0c13970907f1

Thanks
David
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel


Re: [systemd-devel] udev event on usb unpluged

2015-08-26 Thread christophe . jalady
They are not referring to your device, but the machine on which you are
running the rule.
In a terminal/console you run: udevadm monitor
The do the removal itself and see what 'udevadm monitor' says.

this is the problem: the machine which are running the rule is a specific 
robotic device with one usb port and no screen (and no network device).
Either I connect through ssh using the usb cable (eth over usb) or I use the 
usb cable to test my plug/unplug process.
I cannot have a console _and_ test the unplugging at the same time.

Christophe



- Mail original -
De: Robert Milasan rmila...@suse.com
À: systemd-devel@lists.freedesktop.org
Envoyé: Mercredi 26 Août 2015 13:49:44
Objet: Re: [systemd-devel] udev event on usb unpluged

On Wed, 26 Aug 2015 14:43:35 +0200 (CEST)
christophe.jal...@free.fr wrote:

 can you show how you confirmed your script receive the add event,
 or how the script catch the event?
 
 Sure: basically I do echo USB_$ACTION  /tmp/jc.log.
 After plug/wait/unplugged/wait I only see lines USB_add.
 See attached my .rules and .sh scripts.
 
 
 Can you paste run `udevadm monitor`, plug in your cable/device, wait
 for a second, unplug it, and paste the output?
 
 Well, this is the trick: I cannot add a connection to my device _and_
 the usb cable connected to it. So I am not able to monitor the
 process directly from my device. Does the scripts provided are enough
 to help you understanding my problem ?
 
 Christophe

They are not referring to your device, but the machine on which you are
running the rule.

In a terminal/console you run: udevadm monitor

The do the removal itself and see what 'udevadm monitor' says.


-- 
Robert Milasan

L3 Support Engineer
SUSE Linux (http://www.suse.com)
email: rmila...@suse.com
GPG fingerprint: B6FE F4A8 0FA3 3040 3402  6FE7 2F64 167C 1909 6D1A
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel


Re: [systemd-devel] udev event on usb unpluged

2015-08-26 Thread Robert Milasan
On Wed, 26 Aug 2015 15:32:13 +0200 (CEST)
christophe.jal...@free.fr wrote:

 They are not referring to your device, but the machine on which you
 are
 running the rule.
 In a terminal/console you run: udevadm monitor
 The do the removal itself and see what 'udevadm monitor' says.
 
 this is the problem: the machine which are running the rule is a
 specific robotic device with one usb port and no screen (and no
 network device). Either I connect through ssh using the usb cable
 (eth over usb) or I use the usb cable to test my plug/unplug process.
 I cannot have a console _and_ test the unplugging at the same time.

Then login into that machine and run screen, in which you run:
'udevadm monitor -u -p  /tmp/udev.log 21'

Do the necessary removal of the usb cable or whatever, then attach
the usb cable again, login again and check the log.

 
 Christophe
 


-- 
Robert Milasan

L3 Support Engineer
SUSE Linux (http://www.suse.com)
email: rmila...@suse.com
GPG fingerprint: B6FE F4A8 0FA3 3040 3402  6FE7 2F64 167C 1909 6D1A
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel


Re: [systemd-devel] SHM parameters on nspawn containers

2015-08-26 Thread Florian Koch
Hi,

Ohne way is to use an More recent Kernel, with 3.16+ the Kernel defaults for 
These values where changed to  unlimeted

Regards Florian

Von meinem iPhone gesendet

 Am 26.08.2015 um 17:34 schrieb Chris Bell cwb...@narmos.org:
 
 Hello all,
 
 I'm attempting to run GitLab (with postgresql) on a CentOS 7 container with 
 systemd-nspawn. Postgre keeps failing, because it tries to allocate more 
 shared memory than the container seems to allow. I cannot use sysctl to write 
 the kernel.shmmax and kernel.shmall properties, since /sys isn't *real* 
 (sysctl -w fails with 'read-only file system'). I have the values set 
 correctly in the host machine, but they do not seem to propagate/be available 
 to the container. Is there any way I can set (increase) the kernel.shmmax and 
 kernel.shmall values in the container?
 
 Host: systemd 224 on Arch LTS Kernel (3.14.51)
 Guest: systemd 208 on CentOS 7 container
 
 Thanks,
 
 Chris
 ___
 systemd-devel mailing list
 systemd-devel@lists.freedesktop.org
 http://lists.freedesktop.org/mailman/listinfo/systemd-devel
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel


Re: [systemd-devel] udev event on usb unpluged

2015-08-26 Thread Tom Yan
Maybe there's indeed some problem with your kernel or device. I tried
the following rule on my machine:
/etc/udev/rules.d/usb.rules:SUBSYSTEM==usb, RUN+=/usr/bin/bash -c
'echo $ACTION $DEVPATH  /home/tom/test

And here's the test log after I reboot and plug/unplug a usb thumb
drive (2-4):
add /devices/pci:00/:00:1a.0/usb3
add /devices/pci:00/:00:14.0/usb2
add /devices/pci:00/:00:1d.0/usb4
add /devices/pci:00/:00:14.0/usb1
add /devices/pci:00/:00:1d.0/usb4/4-0:1.0
add /devices/pci:00/:00:1a.0/usb3/3-0:1.0
add /devices/pci:00/:00:14.0/usb2/2-0:1.0
add /devices/pci:00/:00:14.0/usb1/1-0:1.0
add /devices/pci:00/:00:14.0/usb2/2-1
add /devices/pci:00/:00:14.0/usb2/2-1/2-1:1.0
add /devices/pci:00/:00:14.0/usb2/2-2
add /devices/pci:00/:00:14.0/usb2/2-2/2-2:1.0
add /devices/pci:00/:00:1a.0/usb3/3-1
add /devices/pci:00/:00:1a.0/usb3/3-1/3-1:1.0
add /devices/pci:00/:00:1d.0/usb4/4-1
add /devices/pci:00/:00:1d.0/usb4/4-1/4-1:1.0
add /devices/pci:00/:00:14.0/usb1/1-13
add /devices/pci:00/:00:14.0/usb1/1-13/1-13:1.0
add /devices/pci:00/:00:14.0/usb1/1-13/1-13:1.1
add /devices/pci:00/:00:14.0/usb1/1-13/1-13:1.2
add /devices/pci:00/:00:14.0/usb1/1-14
add /devices/pci:00/:00:14.0/usb1/1-14/1-14:1.0
add /devices/pci:00/:00:14.0/usb1/1-14/1-14:1.1
add /devices/pci:00/:00:14.0/usb2/2-4
add /devices/pci:00/:00:14.0/usb2/2-4/2-4:1.0
remove /devices/pci:00/:00:14.0/usb2/2-4/2-4:1.0
remove /devices/pci:00/:00:14.0/usb2/2-4

Perhaps you can check the devpath and see if it's actually gone with
`ls` after you unplug the cable/device?

On 26 August 2015 at 21:35, Robert Milasan rmila...@suse.com wrote:
 On Wed, 26 Aug 2015 15:32:13 +0200 (CEST)
 christophe.jal...@free.fr wrote:

 They are not referring to your device, but the machine on which you
 are
 running the rule.
 In a terminal/console you run: udevadm monitor
 The do the removal itself and see what 'udevadm monitor' says.

 this is the problem: the machine which are running the rule is a
 specific robotic device with one usb port and no screen (and no
 network device). Either I connect through ssh using the usb cable
 (eth over usb) or I use the usb cable to test my plug/unplug process.
 I cannot have a console _and_ test the unplugging at the same time.

 Then login into that machine and run screen, in which you run:
 'udevadm monitor -u -p  /tmp/udev.log 21'

 Do the necessary removal of the usb cable or whatever, then attach
 the usb cable again, login again and check the log.


 Christophe



 --
 Robert Milasan

 L3 Support Engineer
 SUSE Linux (http://www.suse.com)
 email: rmila...@suse.com
 GPG fingerprint: B6FE F4A8 0FA3 3040 3402  6FE7 2F64 167C 1909 6D1A
 ___
 systemd-devel mailing list
 systemd-devel@lists.freedesktop.org
 http://lists.freedesktop.org/mailman/listinfo/systemd-devel
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel


[systemd-devel] SHM parameters on nspawn containers

2015-08-26 Thread Chris Bell

Hello all,

I'm attempting to run GitLab (with postgresql) on a CentOS 7 container 
with systemd-nspawn. Postgre keeps failing, because it tries to allocate 
more shared memory than the container seems to allow. I cannot use 
sysctl to write the kernel.shmmax and kernel.shmall properties, since 
/sys isn't *real* (sysctl -w fails with 'read-only file system'). I have 
the values set correctly in the host machine, but they do not seem to 
propagate/be available to the container. Is there any way I can set 
(increase) the kernel.shmmax and kernel.shmall values in the container?


Host: systemd 224 on Arch LTS Kernel (3.14.51)
Guest: systemd 208 on CentOS 7 container

Thanks,

Chris
___
systemd-devel mailing list
systemd-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/systemd-devel


[systemd-devel] Per-connection socket activation not always running service

2015-08-26 Thread John Pitney
To check my understanding of how per-connection socket activation
works, I've tried to make a very simple socket-activated service which
runs any input through md5sum, writes md5sum's output to the journal,
and then sleeps for one second.  What I have works, but only
sometimes.  Frequently, no md5sum output appears in the journal, and
eventually the service goes into a failed state, but I don't see why.

Here are my unit files:

# cat /etc/systemd/system/md5.socket
[Unit]
Description=md5sum socket

[Socket]
ListenStream=8075
Accept=yes

[Install]
WantedBy=sockets.target
# cat /etc/systemd/system/md5@.service
[Unit]
Description=md5sum service

[Service]
ExecStart=/usr/bin/sh -c '/usr/bin/md5sum ; sleep 1'
StandardInput=socket
StandardOutput=journal

Here, I see what to look for in the journal, possibly in some other order.

# for i in $(seq 1 8) ; do echo $i|md5sum ; done
b026324c6904b2a9cb4b88d6d61c81d1  -
26ab0db90d72e28ad0ba1e22ee510510  -
6d7fce9fee471194aa8b5b6e47267f03  -
48a24b70a0b376535542b996af517398  -
1dcca23355272056f04fe8bf20edfce0  -
9ae0ea9e3c9c6e1b9b6252c8395efdc1  -
84bc3da1b3e33a18e8d5e1bdd7a18d7a  -
c30f7472766d25af1dc80b3ffc9a58c7  -

Here, I send the same eight strings to the service.

# for i in $(seq 1 8) ; do echo $i|nc -c localhost 8075  done

Only one of these (number 5) shows up with output in the journal.

# journalctl -b -u md5* --since 20:20:00
-- Logs begin at Sat 2014-11-29 13:38:04 CST, end at Wed 2015-08-26
20:33:19 CDT. --
Aug 26 20:24:55 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55635).
Aug 26 20:24:55 archie-01 sh[24705]: 1dcca23355272056f04fe8bf20edfce0  -

If I try the same for loop a few more times, I got one round of all
eight md5sums appearing in the journal, but in the next round, I get a
few md5sums before I start getting errors.

# journalctl -b -u md5* --since 20:30:00 --no-pager
-- Logs begin at Sat 2014-11-29 13:38:04 CST, end at Wed 2015-08-26
20:36:16 CDT. --
Aug 26 20:35:12 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55643).
Aug 26 20:35:12 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55644).
Aug 26 20:35:12 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55645).
Aug 26 20:35:12 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55646).
Aug 26 20:35:12 archie-01 sh[24752]: 26ab0db90d72e28ad0ba1e22ee510510  -
Aug 26 20:35:12 archie-01 sh[24750]: b026324c6904b2a9cb4b88d6d61c81d1  -
Aug 26 20:35:12 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55647).
Aug 26 20:35:12 archie-01 sh[24756]: 6d7fce9fee471194aa8b5b6e47267f03  -
Aug 26 20:35:12 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55648).
Aug 26 20:35:12 archie-01 sh[24760]: 1dcca23355272056f04fe8bf20edfce0  -
Aug 26 20:35:12 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55649).
Aug 26 20:35:12 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55650).
Aug 26 20:35:12 archie-01 sh[24754]: c30f7472766d25af1dc80b3ffc9a58c7  -
Aug 26 20:35:12 archie-01 sh[24763]: 48a24b70a0b376535542b996af517398  -
Aug 26 20:35:12 archie-01 sh[24766]: 9ae0ea9e3c9c6e1b9b6252c8395efdc1  -
Aug 26 20:35:12 archie-01 sh[24768]: 84bc3da1b3e33a18e8d5e1bdd7a18d7a  -
Aug 26 20:36:16 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55651).
Aug 26 20:36:16 archie-01 sh[24813]: 6d7fce9fee471194aa8b5b6e47267f03  -
Aug 26 20:36:16 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55652).
Aug 26 20:36:16 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55653).
Aug 26 20:36:16 archie-01 sh[24816]: 1dcca23355272056f04fe8bf20edfce0  -
Aug 26 20:36:16 archie-01 sh[24818]: 48a24b70a0b376535542b996af517398  -
Aug 26 20:36:16 archie-01 systemd[1]: Started md5sum service (127.0.0.1:55657).
Aug 26 20:36:16 archie-01 systemd[1]:
md5@14-127.0.0.1:8075-127.0.0.1:55656.service: Failed to run 'start'
task: Transport endpoint is not connected
Aug 26 20:36:16 archie-01 systemd[1]: Failed to start md5sum service
(127.0.0.1:55656).
Aug 26 20:36:16 archie-01 sh[24821]: 84bc3da1b3e33a18e8d5e1bdd7a18d7a  -
Aug 26 20:36:16 archie-01 systemd[1]:
md5@14-127.0.0.1:8075-127.0.0.1:55656.service: Unit entered failed
state.
Aug 26 20:36:16 archie-01 systemd[1]:
md5@14-127.0.0.1:8075-127.0.0.1:55656.service: Failed with result
'resources'.
Aug 26 20:36:16 archie-01 systemd[1]:
md5@13-127.0.0.1:8075-127.0.0.1:55655.service: Failed to run 'start'
task: Transport endpoint is not connected
Aug 26 20:36:16 archie-01 systemd[1]: Failed to start md5sum service
(127.0.0.1:55655).
Aug 26 20:36:16 archie-01 systemd[1]:
md5@13-127.0.0.1:8075-127.0.0.1:55655.service: Unit entered failed
state.
Aug 26 20:36:16 archie-01 systemd[1]:
md5@13-127.0.0.1:8075-127.0.0.1:55655.service: Failed with result
'resources'.
Aug 26 20:36:16 archie-01 systemd[1]:
md5@12-127.0.0.1:8075-127.0.0.1:55654.service: Failed to run 'start'
task: Transport endpoint is not connected
Aug 26 20:36:16 archie-01 systemd[1]: Failed to start md5sum service
(127.0.0.1:55654).
Aug 26 20:36:16 archie-01 systemd[1]: