aladdin134 Posted November 20, 2019 Share Posted November 20, 2019 (edited) I fat fingered a docker path setting up a route to a folder ("/mnt/user/[share name]\"), and now I have a "\" share. this breaks all sorts of things, and I get an error message hidden under the header of every page of my UI, which breaks a lot of functionality. I can't find a way to persistently remove the share in the .ini files in /var/local/emhttp/plugins folder, which is the only such files I can find. I've uninstalled all the dynamix plugins I had, so I don't even know why the template.php is rendering. Warning: syntax error, unexpected $end, expecting TC_DOLLAR_CURLY or TC_QUOTED_STRING or '"' in state/sec.ini on line 273 in /usr/local/emhttp/plugins/dynamix/template.php on line 25 Warning: syntax error, unexpected $end, expecting ']' in state/shares.ini on line 137 in /usr/local/emhttp/plugins/dynamix/template.php on line 29 Warning: syntax error, unexpected $end, expecting TC_DOLLAR_CURLY or TC_QUOTED_STRING or '"' in state/sec_nfs.ini on line 153 in /usr/local/emhttp/plugins/dynamix/template.php on line 30 Warning: syntax error, unexpected $end, expecting TC_DOLLAR_CURLY or TC_QUOTED_STRING or '"' in state/sec_afp.ini on line 267 in /usr/local/emhttp/plugins/dynamix/template.php on line 31 Edited November 20, 2019 by aladdin134 Quote Link to comment
trurl Posted November 20, 2019 Share Posted November 20, 2019 Not entirely clear what you mean by a \ share. Maybe if you could tell us exactly what path you fat-fingered it would clear things up. If you mean the host path was literally \ then that would go away with a reboot as long as the docker is gone and doesn't recreate it, since all the paths at that level are in RAM. If that isn't what you mean explain further. Quote Link to comment
trurl Posted November 20, 2019 Share Posted November 20, 2019 And the templates won't go away simply by removing plugins since the dockerman "plugin" is builtin. Maybe getting us the diagnostics will tell us more. They don't often include much about specific dockers, but they might give us some idea about your actual User Shares. Go to Tools - Diagnostics and attach the complete diagnostics zip file to your NEXT post. Quote Link to comment
aladdin134 Posted November 20, 2019 Author Share Posted November 20, 2019 I added an escape character to the end of the share path and hit enter. here are my diagnostics ivaldi-diagnostics-20191120-1535.zip Quote Link to comment
aladdin134 Posted November 20, 2019 Author Share Posted November 20, 2019 (edited) I now have an entry in my /var/local/emhttp/shares.ini file that looks like ["\"] name="\" nameOrig="\" comment="" allocator="highwater" splitLevel="" floor="0" include="" exclude="" useCache="no" cow="auto" color="yellow-on" free="17603187952" size="0" luksStatus="0" this has persisted through a reboot Edited November 20, 2019 by aladdin134 Quote Link to comment
limetech Posted November 20, 2019 Share Posted November 20, 2019 Diags indicate running 6.8-rc5 - in future please post such reports in the prerelease forum. I moved this topic over there and we'll look into it. Quote Link to comment
trurl Posted November 20, 2019 Share Posted November 20, 2019 55 minutes ago, limetech said: Diags indicate running 6.8-rc5 - in future please post such reports in the prerelease forum. I moved this topic over there and we'll look into it. I was going to respond over there since this is locked now, but it is also locked over there, and most of what I wanted to say have nothing to do with the version anyway, so I will respond here and since @aladdin134 might need to respond to my comments, I am also unlocking this thread. @aladdin134 There is a share anonymized in your Diagnostics as G-------) Could this be the share you mean? It is cache-prefer for some reason and it is all on cache. On the other hand, your system share has files on the array. Probably you created docker and libvirt images before you added cache so they got created on the array instead of cache where they belong. And I see you have given 80G to docker image, and are already using 28G of that. This most likely means you have one or more of your docker applications misconfigured and the app is writing into the docker image instead of to mapped storage like it should. Unless you have something misconfigured, it is extremely unlikely you will need even 20G, and 20G is what I always recommend for the size of docker image. Making it larger doesn't help anything, it just makes it take longer to fill and become corrupted. Do you have any VMs? The best way forward with your docker image is to go to Settings - Docker, disable, delete it then recreate it at only 20G. That will get it on cache where it belongs then Apps - Previous Apps will install your dockers just as they were. But, just as they were isn't the end of it since you have something wrong somewhere. Make sure each place an application is configured to write anything is writing to a mapped path. Common mistakes are not using the same upper/lower case in the application as you used in the docker mappings, or not using an absolute path, beginning with /. Quote Link to comment
limetech Posted November 20, 2019 Share Posted November 20, 2019 34 minutes ago, trurl said: I was going to respond over there since this is locked now Sorry about that - the dumb utility that moves forum posts over to prelease database leaves the new topic locked. It's supposed to lock the source topic not the target topic. Anyway, I moved it because OP is running -rc5. Quote Link to comment
aladdin134 Posted November 20, 2019 Author Share Posted November 20, 2019 I've rolled back to the 6.7 release, and the issue persists. re:docker, I've dropped the dockerfile allocation down (I think there was a bug in one of the images I was using, since fixed, I think most of that storage is my Calibre server media.) that G* share is a share I was using to test gaming performance over 10BaseT, and I was using just the cache to remove the hard disk performance from the equation. I have removed all my VMs previously, and I believe the files have all been removed. The UI for the shares is broken with the same errors, so I can't verify what shares it thinks it has. Quote Link to comment
aladdin134 Posted November 20, 2019 Author Share Posted November 20, 2019 While I can't see the '\' share in the shares config directory, it does show up in the log file, breaking my samba service Nov 20 14:02:34 Ivaldi emhttpd: Starting services... Nov 20 14:02:34 Ivaldi emhttpd: shcmd (16659): /etc/rc.d/rc.samba restart Nov 20 14:02:36 Ivaldi root: Starting Samba: /usr/sbin/nmbd -D Nov 20 14:02:36 Ivaldi root: /usr/sbin/smbd -D Nov 20 14:02:36 Ivaldi root: /usr/sbin/winbindd -D Nov 20 14:02:37 Ivaldi emhttpd: error: put_config_idx, 609: Invalid argument (22): fopen: /boot/config/shares/TV\ Shows.cfg Nov 20 14:02:37 Ivaldi emhttpd: error: put_config_idx, 609: Invalid argument (22): fopen: /boot/config/shares/\.cfg Nov 20 14:02:37 Ivaldi emhttpd: Starting services... Nov 20 14:02:37 Ivaldi emhttpd: shcmd (16669): /etc/rc.d/rc.samba restart Quote Link to comment
limetech Posted November 20, 2019 Share Posted November 20, 2019 I can recreate this bug. To get rid of this, using terminal window or console you should be able to see the share names like this: v /mnt/user/ Maybe you will see a directory that looks like this: \\ If so you can delete it like this: rmdir "/mnt/user/\\" Might need to stop/start array afterwards. Quote Link to comment
aladdin134 Posted November 20, 2019 Author Share Posted November 20, 2019 that solved it, thanks! Quote Link to comment
lsaranto Posted December 10, 2019 Share Posted December 10, 2019 Thanks! I just had the same error. I had run a script with too few double quotes. In the end I had a share with a backslash in its name. Removed the share and everything is fine. 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.