As most of your work is using these SMB shares, have you experienced any failures of access?
SMB has been around for decades, so should be debugged, I would think. But we all know computers and networks are not perfect, especially when we need them to be perfect.
Perhaps others have a view or experience, but as DEVONthink does not modify files and only reads the files to create indexes, I can’t think of anything worse than simply the file on the remote share is unreadable or something, or after it’s been indexed then due to a technical failure when you want to look at a file it can’t be loaded, or something.
Me, I’d just “have at it”. Perhaps create a new database for it all.
The smb shares do unmount from time to time, as I can only access them via VPN when not docked.
It is also my immediate impulse to just have at it. I have already indexed my personal smb drive at work and have not had issues. I suppose that the destructive potential of DT indexing is limited to fairly deliberate deletions. But as I now want to try the indexing on truly shared drives, I just wanted to check with more savvy users than myself before potentially setting a lot of my institution’s operational history on fire.
I don’t know that I’m more savvy or not, but I can’t think why not to do it. Doesn’t mean risk-free, I guess.
Carefully read the DEVONthink Manual about Indexing (page 57 of version 3.8.7 manual), and if you delete indexes, read the dialog boxes carefully.
Hopefully you do not have DELETE permissions on other people’s SMB shares.
And that your institution has backups. Not sure how big your institution is, one of my old nameless institutions were gung-ho about ensuring backups and all that, but their bureaucracy all but prevented timely restore of files (sometimes accidenty deleted).
Thank you for your reply. I absolutely agree with the blog post that using DT as an excuse to be disorganized is not a great idea. In the case at hand, I have inherited someone else’s junk drawer – one that is no longer in use, but also one that I don’t want to change either.
It is about 2GB of mess that I would like to render searchable. For the indexing process I would be on the same gigabit Ethernet connection. Whether the database corrupts is not a big problem. I can always throw it away and start again. I just don’t want to perform any destructive actions on institutional files that, although in a state of ruins, seem to be the only history we have of certain processes. And with regards to the hope of @rmschne, I actually do have DELETE permissions, although I would prefer not to.
2GB not really that much in today’s world. Maybe COPY it from the server to your local drive, import to DEVONthink and work on there? Avoids risk of accidental delete of “institutional jewels” on the server and ameliorates any risks of DEVONthink running off index files that are “remote”.