snowboardjoe Posted September 14, 2017 Share Posted September 14, 2017 (edited) Was starting to review steps needed to migrate to PRO and was so happy to see this thread here. A few questions, though: 1. I'm backing up 11TB and Code42 has already indicated it can't migrate anything over 5TB and will need to perform a full backup from scratch. I assume there is no need for me to migrate any data then? Or do I still need to do copy "/mnt/user/appdata/CrashPlan" to CrashPlanPRO? 2. Long ago I had to modify the code/config to avoid deduplication by CP as it bogged down multi-terabyte backups. Will I need to make that modification again with PRO? I need to find in my notes how I did this long ago. With dedupe turned on my backup never would not have finished as it was too aggressive. Edited September 14, 2017 by snowboardjoe Quote Link to comment
FreeMan Posted September 14, 2017 Share Posted September 14, 2017 (edited) On 8/25/2017 at 0:56 PM, JustinAiken said: Okay, was successfully able to migrate from the other container without adoption! cd /mnt/cache/appdata cp -r CrashPlan/ crashplan-pro chown -R nobody:users /mnt/user/appdata/crashplan-pro/ cd crashplan-pro/ mv id/ var/ rm -r cache rm -r bin rm -r log All docker template values are default, except for: - config -> /mnt/user/appdata/crashplan-pro - data -> /mnt/user (To match old container) So far, so good. I followed these steps, CPP came right up and it's currently synchronizing block information. It says all my destination points are missing, though, but I'll wait for the sync to finish and see what it says then. The worrying part was dealing with the code-42 website. I entered all the info, clicked the button and basically nothing... it went back to the start of the enrollment screen, and now the Admin panel has a scary red warning that my subscription is about to expire and I need to re-enroll now!!! It says the subscription is good through March 2018, though, although with my extra 60 days it should be beyond that. If things don't clear up by tomorrow evening, I'll jump on chat with them & get it straightened out. Thanks to @Djoss for getting this container working so quickly and to @JustinAiken for the simple directions! UPDATE: For those who have large backups, don't forget to allocate more memory to CrashPlan. Double click on the little business building icon in the top right corner then type java mx xxx, restart Where xxx is the amount of memory you want to assign in MB. Their instructions are listed here. They recommend 1 GB of memory for every TB of backup or million files. Edited September 14, 2017 by FreeMan Quote Link to comment
Djoss Posted September 15, 2017 Author Share Posted September 15, 2017 1 hour ago, snowboardjoe said: 1. I'm backing up 11TB and Code42 has already indicated it can't migrate anything over 5TB and will need to perform a full backup from scratch. I assume there is no need for me to migrate any data then? Or do I still need to do copy "/mnt/user/appdata/CrashPlan" to CrashPlanPRO? You can effectively start with an empty appdata. No need to keep your previous config. 1 hour ago, snowboardjoe said: 2. Long ago I had to modify the code/config to avoid deduplication by CP as it bogged down multi-terabyte backups. Will I need to make that modification again with PRO? I need to find in my notes how I did this long ago. With dedupe turned on my backup never would not have finished as it was too aggressive. Yes you need to redo it. Make sure to read the article from Code42 about his (unsupported) change: https://support.code42.com/CrashPlan/4/Configuring/Unsupported_changes_to_CrashPlan_de-duplication_settings Quote Link to comment
Djoss Posted September 15, 2017 Author Share Posted September 15, 2017 1 hour ago, FreeMan said: UPDATE: For those who have large backups, don't forget to allocate more memory to CrashPlan. Double click on the little business building icon in the top right corner then type java mx xxx, restart Where xxx is the amount of memory you want to assign in MB. Their instructions are listed here. They recommend 1 GB of memory for every TB of backup or million files. Much more easier to set the memory amount using the dedicated container's setting! Quote Link to comment
FreeMan Posted September 15, 2017 Share Posted September 15, 2017 2 minutes ago, Djoss said: Much more easier to set the memory amount using the dedicated container's setting! True! I saw that but it didn't even register... Thanks, I've made the change there, since it probably would have been overwritten next time the container was updated. Quote Link to comment
FreeMan Posted September 15, 2017 Share Posted September 15, 2017 11 minutes ago, FreeMan said: Thanks, I've made the change there, since it probably would have been overwritten next time the container was updated. And... now I've unset the change there. It wouldn't start properly with it set in both places. Quote Link to comment
Djoss Posted September 15, 2017 Author Share Posted September 15, 2017 You mean when set in container's setting the container is not starting? Can you provide the content of /mnt/user/appdata/CrashPlanPRO/bin/run.conf? Quote Link to comment
FreeMan Posted September 15, 2017 Share Posted September 15, 2017 I initially set it (to 2560) by using the "click on the app" method I described and it was fine. Then, you mentioned the setting in the container settings, so I edited the settings and set it there to "2.5G". When I restarted the container, I couldn't get the WebGUI to come up. I cleared the field in the container settings and restarted again, and it came up. Contents of run.conf: SRV_JAVA_OPTS="-Dfile.encoding=UTF-8 -Dapp=CrashPlanService -DappBaseName=CrashPlan -Xms20m -Xmx2560m -Dsun.net.inetaddr.ttl=300 -Dnetworkaddress.cache.ttl=300 -Dsun.net.inetaddr.negative.ttl=0 -Dnetworkaddress.cache.negative.ttl=0 -Dc42.native.md5.enabled=false" GUI_JAVA_OPTS="-Dfile.encoding=UTF-8 -Dapp=CrashPlanDesktop -DappBaseName=CrashPlan -Xms20m -Xmx512m -Dsun.net.inetaddr.ttl=300 -Dnetworkaddress.cache.ttl=300 -Dsun.net.inetaddr.negative.ttl=0 -Dnetworkaddress.cache.negative.ttl=0 -Dc42.native.md5.enabled=false" After restarting, it will come up, but it cannot connect to the server, and I see this in the settings: I'll stop fiddling with it until I get some advice. (Which has been quite good so far, if only I'd slow down some!) Quote Link to comment
Djoss Posted September 15, 2017 Author Share Posted September 15, 2017 My guess is that the value you set failed to be replaced in run.conf. If you look at the container's logs, you will probably see an error that prevent the container to start. One thing you can try is to stop the container and remove run.conf. The set the memory amount in the container settings and start the container. Quote Link to comment
FreeMan Posted September 15, 2017 Share Posted September 15, 2017 I looked at run.conf and it had "-Xmx1024m", so I deleted run.conf, set memory to "2.5G" and restarted. Again I couldn't bring up the UI. Stopped the container, cleared the memory setting from the WebGUI settings. It didn't start the container, so I edited the run.conf to read "-Xmx2560m" and started the container. The UI will come up now, but it's still not starting the backup, and the internal address is still showing 0.0.0.0:4242. I could uninstall/reinstall the whole container if you think that would be a good next step. Quote Link to comment
Djoss Posted September 15, 2017 Author Share Posted September 15, 2017 I think the issue is with the "2.5G" value. Not sure CP supports fraction. Try to set "2560m" instead. The backup issue is a separate thing. I think Code42 is doing some maintenance. It's also not working on my side. Quote Link to comment
Shadowrunner Posted September 15, 2017 Share Posted September 15, 2017 9 hours ago, snowboardjoe said: Was starting to review steps needed to migrate to PRO and was so happy to see this thread here. A few questions, though: 1. I'm backing up 11TB and Code42 has already indicated it can't migrate anything over 5TB and will need to perform a full backup from scratch. I assume there is no need for me to migrate any data then? Or do I still need to do copy "/mnt/user/appdata/CrashPlan" to CrashPlanPRO? 2. Long ago I had to modify the code/config to avoid deduplication by CP as it bogged down multi-terabyte backups. Will I need to make that modification again with PRO? I need to find in my notes how I did this long ago. With dedupe turned on my backup never would not have finished as it was too aggressive. I switched over and my full 17TB migrated without having to re-upload anything, I'm not alone in this either. SR Quote Link to comment
snowboardjoe Posted September 16, 2017 Share Posted September 16, 2017 I'm still trying to see if I can takeover my existing 11TB of data. CPPro still shows the data is there after the migration and I've been struggling to get my new client to attach to it to keep from spending days/week backing it up all over again. I went through the takeover steps, but it still wants to start over from scratch. When I first got it initialized it shows all my shares were missing and guessing this was due to the path changes (using shares instead of storage and using flash instead of boot). The instruction say to just reselect the shares (not remove the old ones), but it still insists on backing up everything all over again. Is there any way around this? Is there something I should be mapping differently to make it all work? I thought the client would have figured out that data had moved and would not back it up again. Quote Link to comment
FreeMan Posted September 16, 2017 Share Posted September 16, 2017 Yeah, well, after a flawless conversion from gfjardim's docker to this one, there was one minor issue... I screwed up my setting for the backup path so the backups didn't start. Then I started messing around trying to figure out why everything was listed as missing and started a prune. When that finishes, I think I'll end up with an empty backup set at code42 and get to start over. Totally my fault, but be sure to learn from my mistake! Quote Link to comment
francrouge Posted October 7, 2017 Share Posted October 7, 2017 Can someone help me. Since i reboot my unraid pc the crashplan engine wont start. I got the crashplan pro etc. Quote Link to comment
Djoss Posted October 7, 2017 Author Share Posted October 7, 2017 Do you have the container’s log? Quote Link to comment
francrouge Posted October 7, 2017 Share Posted October 7, 2017 1 minute ago, Djoss said: Do you have the container’s log? I just reinstalled the app and its working. But if i restart it it is not working 06.pdf Quote Link to comment
Djoss Posted October 7, 2017 Author Share Posted October 7, 2017 Nothing special in container's log. Under the container's appdata (/mnt/user/appdata/CrashPlanPRO/log/), you can look at these log files: service.log.0 engine_error.log engine_output.log Quote Link to comment
francrouge Posted October 7, 2017 Share Posted October 7, 2017 4 minutes ago, Djoss said: Nothing special in container's log. Under the container's appdata (/mnt/user/appdata/CrashPlanPRO/log/), you can look at these log files: service.log.0 engine_error.log engine_output.log service.log.0 engine_error.log engine_output.log Quote Link to comment
Djoss Posted October 7, 2017 Author Share Posted October 7, 2017 It seems to be the problem: java.io.IOException: IO error: /usr/local/crashplan/conf/adb/000005.ldb: No such device Is your appdata correctly seen from the container? Can you run these commands on the server: docker exec CrashPlanPRO ls -l /config/ docker exec CrashPlanPRO ls -l /config/conf Quote Link to comment
francrouge Posted October 7, 2017 Share Posted October 7, 2017 4 minutes ago, Djoss said: It seems to be the problem: java.io.IOException: IO error: /usr/local/crashplan/conf/adb/000005.ldb: No such device Is your appdata correctly seen from the container? Can you run these commands on the server: docker exec CrashPlanPRO ls -l /config/ docker exec CrashPlanPRO ls -l /config/conf Quote Link to comment
Djoss Posted October 7, 2017 Author Share Posted October 7, 2017 If you stop and then start the container (without reboot), do you have the issue? If you manually stop the container before rebooting, do you have the issue? Quote Link to comment
francrouge Posted October 8, 2017 Share Posted October 8, 2017 no I have to delete everthing and reinstalled the app. thx Quote Link to comment
Lebowski Posted October 8, 2017 Share Posted October 8, 2017 Just wanted to ask about restoring a file. I wanted to test out a restore via the docker. I was pointing a file to restore it to a share and it said it didn't have write access. I tested a various amount of shares inc a public share. Even back to the original location. Write permission error. Can anyone else replicate this? I am using default mappings. My backups are working fine. Its more of a restore withing the docker issue I am having. Cheers, Quote Link to comment
Djoss Posted October 8, 2017 Author Share Posted October 8, 2017 By default, /storage is read-only. This is to make sure that data to backup is not touched/modified. If you want to restore, you can create another mapping (add a "Path" in container's setttings) with read/write permissions and restore to this location. You can also change the permissions of /storage if you prefer. 1 Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.