Crashplan 4.3.3 says it is running, but I can't connect

[pcloadetter]

Through a long convoluted procedure, I think Crashplan is running. But I can't connect.

cat var/log/crashplan/engine_output.log | grep CPV

CPVERSION = 4.3.3

/usr/pbi/crashplan-amd64/share/crashplan/bin/CrashPlanEngine status

CrashPlan Engine (pid 86259
86260
86261
86263
86264
86265
86266
86267) is running.

service crashplan status

/usr/local/etc/rc.d/crashplan: WARNING: no shebang line in /usr/bin/cpuset
crashplan is not running.

/usr/pbi/crashplan-amd64/share/crashplan/conf/my.service.xml

<serviceUIConfig>
<serviceHost>0.0.0.0</serviceHost>
<servicePort>4367</servicePort>

/usr/pbi/crashplan-amd64/share/crashplan/log # netstat -na | grep LISTEN

tcp4 0 0 *.4366 *.* LISTEN
tcp4 0 0 *.4367 *.* LISTEN
tcp4 0 0 192.168.10.231.12351 *.* LISTEN

/var/lib/crashplan/.ui_info

4367,<authkey>,0.0.0.0

So I would think my port is 4367.
When I try a telnet to 4367, it hangs, but doesn't disconnect.
When I try a telnet to 4366

telnet 192.168.10.231 4366
cA-18782|com.code42.messaging.security.SecurityProviderReadyMessage"▒p▒▒▒▒▒)9

My Windows files..
ui.properties

serviceHost=192.168.10.231
servicePort=4367

.ui_info

4367,<authkey>

So what am I doing wrong?

 

[dlavigne]

What error do you get when you try to connect? Anything in /var/log/messages of the plugin's jail shell?

 

[pcloadetter]

The Crashplan client screen just hangs.

/var/log/messages

/usr/local/etc/rc.d/crashplan: WARNING: no shebang line in /usr/bin/cpuset

 

[dlavigne]

Sounds like a problem with the startup script. Does /usr/local/etc/rc.d/crashplan look like this: https://svnweb.freebsd.org/ports/he...iles/crashplan.in?revision=371372&view=markup ?

 

[pcloadetter]

Nope. Only the first few lines. Mine is this:

Code:

#!/bin/sh
#
# $FreeBSD$

# PROVIDE: crashplan
# REQUIRE: DAEMON

. /etc/rc.subr

name="crashplan"
rcvar=${name}_enable

command="/usr/bin/cpuset -l 0 /usr/pbi/crashplan-amd64/share/crashplan/bin/CrashPlanEngine"

pidfile="/var/run/CrashPlanEngine.pid"

start_precmd="install -d /var/log/crashplan"
start_cmd="${command} start"
stop_cmd="${command} stop"

load_rc_config ${name}
run_rc_command "$1"

 

[Thirtybird]

Just throwing my hat into this as having the same problem. I've gone through and with the jail shut down I've re-set the my.service.xml and .ui_info back to the values they were before the plugin updated from 4.3.0 to 4.3.3, and I still cannot get the gui interface to come back up. The crashplain jail is listening on 4243 and an inbound telnet to that port from the gui machine does connect.

Checking out code42's website, 4.3.3 isn't even listed as a home user version yet - they only have 4.3.0 listed, and that's the version available for download for Windows.

EDIT: It appears the .ui_info is being rewritten everytime I restart either the jail or the service - unknown if that is related.

EDIT 2:I took the contents of the .ui_info file on my FreeNas and replaced the contents of the corresponding file on my windows client and was then able to connect. Not ideal if you need to connect to multiple machines from there, but I only connect to the local and the freenas, so it works for me for now. I'll have to see what triggers the rewrite of the .ui_info file on the freeNAS

 

[pcloadetter]

Just curious, for the .ui_info file, do you include the port as part of you key copied or just the data after it? I don't believe that's the issue here, but I'm curious all the same.

 

[Thirtybird]

The port wasn't being overwritten, 4243 was still there, so I really only needed to copy the key part.

 

[dlavigne]

If you copy the existing rc script to crashplan.orig, save https://svnweb.freebsd.org/ports/he...an/files/crashplan.in?revision=371372&view=co to /usr/local/etc/rc.d/crashplan, and run service crashplan start, does it fix it? (you may have to set the execute bit on the new script so it matches the perms of crashplan.orig)

 

[Raven9810]

I am having the exact same issues as pcloadetter. I can confirm that mine appears to be running, looking at the CrashPlanEngine status, but it is NOT backing up files. On mine, it appeared that the service port had changed for some reason, from 4243 to 4255. I made changes to the UI.info file on the server to match that of the client installed on my Windows machine. I also ensured that the my.service.xml lines for Location and Service port matched that of the client. Still not backing up. I have also tried the opposite, taking what is in these files and matching the client to them. Still doesn't backup.

I also attempted what dlavigne suggested and it did not fix the issue.

I may just uninstall the plugin and re-install. But it is getting frustrating that even though nothing is being changed on the server, Crashplan will all of a sudden stop working like this. I just went through something similar in July. Crashplan was working for a while since last October, and then just stopped working. I eventually was able to get that working again, but those fixes that helped in July are not helping now.

I am hoping someone who is going through this same issue, posts information if they figure out the fix.

 

[dlavigne]

Bummer...

 

[pcloadetter]

If you copy the existing rc script to crashplan.orig, save https://svnweb.freebsd.org/ports/he...an/files/crashplan.in?revision=371372&view=co to /usr/local/etc/rc.d/crashplan, and run service crashplan start, does it fix it? (you may have to set the execute bit on the new script so it matches the perms of crashplan.orig)

I copied the original to crashplan.orig and then edited the existing one.

root@crashplan_1:~ # service crashplan start
sed: %%PREFIX%%/share/crashplan/bin/run.conf: No such file or directory

 

[pcloadetter]

There is a mention of 4.3.3 on their site now.
http://support.code42.com/Release_N..._Cloud_Version_4.3.0#Maintenance_Update_4.3.3. Basically, nothing that I can see of relevance.

 

[cstathoulis]

Nope. Only the first few lines. Mine is this:

Code:

#!/bin/sh
#
# $FreeBSD$

# PROVIDE: crashplan
# REQUIRE: DAEMON

. /etc/rc.subr

name="crashplan"
rcvar=${name}_enable

command="/usr/bin/cpuset -l 0 /usr/pbi/crashplan-amd64/share/crashplan/bin/CrashPlanEngine"

pidfile="/var/run/CrashPlanEngine.pid"

start_precmd="install -d /var/log/crashplan"
start_cmd="${command} start"
stop_cmd="${command} stop"

load_rc_config ${name}
run_rc_command "$1"

EDIT

Could you please clarify something for me. The code above may be the solution but this is what I would like to do using a POST INIT script. I would like to create a script which will, copy the "correct/modified"

/usr/pbi/crashplan-amd64/share/crashplan/bin/run.conf
/usr/pbi/crashplan-amd64/share/crashplan/conf/my.service.xml

in their respective paths and then run the crashplan service .

Each time I reboot the FreeNas, I lose this and have to keep doing it manually.

Many Thanks

 

[pcloadetter]

I don't think I'm the one to clarify that for you, sorry. I'm just as stumped.

 

[Mark Hodgkinson]

I had the same problem, to get it working again I had to revert back to forwarding the port via SSH instead of connecting the CP client directly to the host over the network how I used to have it working.

For Windows I did:

1 - Install OpenSSH
http://www.mls-software.com/opensshd.html

2- In the CP client software folder, open the C:\Program Files (x86)\CrashPlan\conf\ui.properties (it may have a username.properties if you install it per user)
Change the contents to below:

#Fri Dec 09 09:50:22 CST 2005
#serviceHost=192.168.0.21
servicePort=4200
#servicePort=4250
#pollerPeriod=1000 # 1 second
#connectRetryDelay=10000 # 10 seconds
#connectRetryAttempts=3
#showWelcome=true

#font.small=
#font.default=
#font.title=
#font.message.header=
#font.message.body=
#font.tab=


I.e. Just add port 4200 enabled, it must be 4200 for this setup, not any other number

3 - Open C:\ProgramData\CrashPlan\.ui_info

Change the contents to:

4200,XXXXXXXXXXXXXXXXXXXXXXX,127.0.0.1

The XXXXXXXXXXXXXXXXXXXXXXX number comes from the /var/lib/crashplan/.ui_info on your CP server but you must leave the 4200, and ,127.0.0.1 the same, it doesn't matter what those values are in the file on your server, ignore them!

4 - In OpenSSH run the following command:

In C:\Program Files (x86)\OpenSSH\bin\
run:
ssh -L 4200:localhost:4243 root@192.168.0.21

Where 192.168.0.21 is the hostname of your Crashplan server
The 4243 two lines above came from the value in my .ui_info on the server (just in case yours is different, it is the standard port number though)

See Point 6 if you can not login via SSH

5 - Login and then load the Crashplan client and hopefully it works for you now as well!


6 - If you can not login to your jail via SSH then do the following in the jail:

Edit:
/etc/rc.conf
to enable:
sshd_enable=YES

and edit:
/etc/ssh/sshd_config
Change the following:
PermitRootLogin yes

On the shell then type:
passwd
To set a root password:

Then run:

service sshd restart

Go to point 5 again.

 

[pcloadetter]

I deleted and reinstalled the CP plugin. Now the whole thing is out of wack. Not sure if I need to do a reboot or something to clear out the cruft. But I can't even test anymore.

 

[sdgenxr]

Mark, thanks for the detailed instructions. However, I'm a little confused on step #4 as to exactly where we should run those commands. Do we launch OpenSSH and run those commands, does it have to be done in OpenSSH or can we use another client such as PUTTY?

Not sure if it's related but I've also noticed that Crashplan no longer autostarts at reboot.

 

[Phlogi]

FYI: There might be some useful debug information in the jail:

Code:

/usr/pbi/crashplan-amd64/share/crashplan/log/

The file of interest is service.log.0

While connecting you can watch it with:

Code:

tail -f /usr/pbi/crashplan-amd64/share/crashplan/log/service.log.0

Now you can see what happens while you try to connect:

Code:

[10.02.15 11:25:01.377 INFO    Sel-UI-A             com.backup42.service.ui.UIController    ] UISession Created - 709727073838178422
[10.02.15 11:25:01.832 INFO    MQ-UI-1              com.backup42.service.ui.UIController    ] UserActionRequest: StatusQueryMessage[709727073838178422] CPD|91782d2e-8085-474a-8816-232968f6561c|Phlogi|/Users/Phlogi|
[10.02.15 11:25:01.832 INFO    MQ-UI-1              com.backup42.service.ui.UIInfoUtility   ] Loaded address=127.0.0.1, port=4243, and token from location=/var/lib/crashplan/.ui_info
[10.02.15 11:25:01.915 INFO    MQ-UI-1              com.code42.logging.AppLogWriter         ] WRITE app.log in 64ms
[10.02.15 11:25:01.925 INFO    MQ-UI-1              com.backup42.service.ui.UIController    ] UserActionResponse: StatusResponseMessage@9704897[ session=709727073838178422 ][ User@-1171233680[ userId=,  ], AuthorityLocation@13465049[ location=central.crashplan.com:443, hideAddress=false, isLocked=false ], orgType=CONSUMER, upgrading=false, UpdateLicenseMessage[] [ authenticated=true, updateConfigPassword=false, username=phlogi1@gmail.com, securityKeyType=PrivatePassword, dataKeyExists=true, secureDataKeyExists=true, secureDataKeyQA=null, authorizeRules=AuthorizeRules [hashPassword=true, ldap=false, minPasswordLength=5, usernameIsAnEmail=true, deferredAllowed=false], blocked=false ], locale=en_US, websiteHost=https://www.crashplan.com, defaultRestoreFolder=/usr/pbi/crashplan-amd64/share/crashplan/conf, os=Linux ]
[10.02.15 11:25:03.877 INFO    Factory$Notifier-UI0 com.backup42.service.ui.UIController    ] UISession Ended after less than a minute - 709727073838178422
[10.02.15 11:25:03.878 INFO    Factory$Notifier-UI0 backup42.service.backup.BackupController] UI:: AUTO IDLE... lowBandwidth=0 B/s, activeThrottleRate=80
[10.02.15 11:25:03.878 INFO    Factory$Notifier-UI0 com.backup42.common.command.CliExecutor ] RUN COMMAND: auto.idle  >  CommandResult@1341464[ auto.idle, success=true, result=null ]

So it seems that the connection with the tunnel works properly, but the GUI just keeps hanging...