Spectrum

Now that I have my unRAID box set up and stable, I'm setting up some rsync scripts to back up data from my primary workstation. I have a pretty extensive photo library of Canon raw files (that keeps growing )and want to keep them mirrored on the unRAID box for security. In the past, I have almost lost some images due to bit-rot. Example: I pulled up a photo I knew was good and it had fluorescent streaks and patches. I pulled a copy from my offsite backup and it was fine. My concern here that if I get bit-rot on my primary workstation, I don't want it mirrored to the unRAID box. As a test I did the following: Created a 15MB binary file on the unRAID box Copied the file to my windows box Changed 1 bit in the file Copied it back to the unraid box Used touch to make the modified file's timestamps match the original file's Copied the file back to my windows box Ran rsync with the modified file as the source and pointed the dest to the location of the original file on the unRAID box Doing a cmp on the modified file on the windows box and the original file on the unRAID server showed the files as being different after the rsync, so rsync did not copy the "bad" file over the "good" file. Based on this I think it is safe to proceed with scripting and forgetting, but I wanted to check in here to see if anyone else had any thoughts or insights.

Glad to confuse, I mean help you If you see anything that is unclear or needs improvement, let me know. That was a pretty massive data dump and I know there are mistakes in there somewhere!

They can but feel flimsy if you do. I put 4 screws in per cage just to firm it up a bit.

@Raj Joe L. has edited the wiki and he put in only run 4 at once. I initially put 6 in with a comment that I needed the actual# from Joe and he took care of it @Userpaul It's been there as long as that section has lived But I've made it a bit more noticeable now.

Sounds like the Preclearing with Screen section needs some more work. And Joe L. already did it, thanks! On US keyboards the pipe symbol "|" is usually paired with the "\" key (requiring shift to get to it) and looks like 2 bars, one on top of the other instead of a single bar as shown when typed. Wikipedia has some more info on the ambiguities if you are interested in drowning in information. If you have a monitor and keyboard attached to your server you may have an easier time preclearing from the console's V-terms as described in the section Preclearing Without Screen.

I'm 99% sure shares are top level only. I've been in the docs extensively for the past few days setting up my system and working on that tutorial. With that said, someone may pipe up and prove me wrong With that out of the way, for what you have shown you would need shares for Media, Business1, Business2, and Personal. Anything inside those shares would just be a folder/directory. The upside is that it's easy to configure, the downside is that you can only set permissions on the shares themselves and not at the folder level inside the shares, but that's beyond the scope of your question

I "finished" the tutorial tonight There are still a few things that need to be done though. Thanks GBH2 for the info on the mail setup. That has been assimilated and posted! Joe L. if you are paying attention, where would you draw the line on the max number of preclears to run at once? I realize this is hardware dependent, but what would be a good number for an "average" system? I cheated on the User Shares section and pretty much just listed the options and linked to the un-official manual. I want to think about the verbiage before I write out info on allocation and split leveling, but if someone wants to beat me to it and make it purty go right ahead! After thinking about this, I decided to leave those sections in. Yes they are kind of like troubleshooting steps, but they don't take long and I had rather have a noob type ifconfig and see that there is no ip address rather than try to troubleshoot name resolution if their Windows boxen won't find //tower on the network. Also grepping out that section of dmesg forces them to take a look at their hard drive speeds and potentially fix jumpers if they have a SATAII drive locked in SATAI mode, etc. Of course maybe I am off base and am over geeking here? It's been known to happen Now what we need is a complete neophyte that has no knowledge of Linux or Windows networking to step through this un-aided and point out all the errors!

WOW insomnia FTW tonight!!! I'm in for one too at $20. Gives me a good excuse to order that non-blue LED case fan I need too

Did my instructions for how to get the files and copy them over make sense? I thought about walking through using wget from the console but decided most people prefer the comfort of a gui and want a little to do with command line as possible. Hmm didn't use the mail stuff myself. I'm at my system enough it was no problem for me to telnet in and check the screen session to see if it was done. If someone doesn't feel up to editing the wiki but doesn't mind writing it up, I'll be glad to format it and include it on the page.

Good points all! I am going from my (pathetic) notes that I kept with this in mind as I was setting up my server over the past few days. If you feel like making changes before I get to it, it is a wiki so go for it! But I will address these when I jump back on it in the next day or so if no one else does. Setting your DNS server to your gateway should work with most modern routers, but I have had some in the past that would refuse to pass dns traffic they received on port 53. They assumed it was destined for them since the packet had their address as dst but they had no daemon running on port 53 so requests would just time out. Most modern routers will just forward requests on port 53 to whatever DNS servers they have in their list and change the dst address in the packet then rewrite the response packet and send it back to the system inside the network that requested it. I wouldn't count on that working in all configs, but most modern routers should take care of it

Being a newbie UnRAID user and somewhat inspired by jwcolby's posts in this thread I decided to start a tutorial that would help a new user get from the "OK I have a working flash drive, now what" stage to a usable array. I agree with jwc that there is (was ) a gap in the documentation and there was no 'one stop shop' of the information required to get from that working flash drive state to a working array, so that is the slant I took in the wiki article. It's not meant to be the end-all be-all but a step by step tutorial on getting things set up that doesn't require hunting down different wiki posts, forum threads, google code sites, etc. There are links to all of those places, but my goal is to minimize new user's having to search for the right ones and give them details on what they are looking for. I made it in the first boot section of the getting started with UnRAID as that seemed most appropriate. Agree/disagree? Direct link to the article is http://lime-technology.com/wiki/index.php?title=Configuration_Tutorial I've gotten through to the preclearing stage and thought I would go ahead and commit the changes and ask you guys what you think and to take a look and fix my mistakes

Just make sure you have the switch connected to the internal ports of the router, not the WAN/internet ports. Router Switch +-------------------+ +---------------+ | | | | | W 0 1 2 3 | | 0 1 2 3 | +---|---|---|---|---+ +---|---|---|---+ | | | To --+ +---------------+ Internet

Booo! I just ordered that for my first build and it was delivered about an hour ago. Oh and hi from a newb