My Synology DS212+ NAS went wonky today. Wonky seems like an appropriate technical term. It had stopped indexing some photos I uploaded so I took the usual troubleshooting step and rebooted. Then things went down hill from there.
After the reboot, and logging on with the admin ID, I would keep getting the initial Welcome Wizard although it wouldn’t let me actually do anything and run through the wizard. Despite this, I could access the file shares.
After the usual browser troubleshooting steps I hadn’t made any progress. I found that I could access the server from my iPhone if I used DSM Mobile. But as soon as I would try the full DSM I’d get the wizard.
DSM Mobile gave me enough access so that I could give my regular ID system administrator privileges. Once I did that I could logon to the full website with my regular ID but with greater access to check things out. That’s when I noticed that the DSM Upgrade gave me a out of disk message. A full system volume sure would explain a lot although there’s not much I could do from the GUI.
First I enabled SSH:
Then I used terminal to SSH into the Synology NAS as root. The root password is the same password given to the Admin ID.
>ssh firstname.lastname@example.org (IP address of Synology NAS)
Then I worked my way down to find out which directory is using too much and then the large files:
>du -hs *
This will list all the directories and their sizes. I change into the largest directory (mine was nearly 2GB and was var). I switched into the directory and executed ds -hs * again. Eventually I found I had two 800+ MB files in /var/log/httpd. Both were archived logs so I deleted them as follows:
I still had a much smaller sys-cgi_log file so it seemed safe to delete those two.
I rebooted after deleting the files.
I still had to run through the welcome wizard when I logged on with admin. I simply picked the option to “skip” any configuration and was brought to the DSM desktop. My user specific desktop settings were gone but all files and services are there.
The monitoring app and widget can’t connect to the service so won’t run. This is a minor annoyance. A search of the forums shows a re-install as a solution. If this is the only problem I have, I’ll wait to see if the next patch fixes the problem.
What caused the log to grow so large still needs some research. But for now I’ll monitor their size.
July 2014 brought the end of an era that began in January of 2008. I shut down my Windows Home Server. Except for a brief two month fling with an Ubuntu home server I’ve had a Windows Home Server running for the last six and a half years. There’s nothing replacing it. Although, an existing Synology NAS takes over some duties.
My Windows Home Server started with two small drives on a HP Windows Home Server version 1. It grew to a home built box with over 20 TB of disk by the time WHS 2 was released. Eventually it began to shrink and by the time I shut it down it was an HP MicroServer with four 3 TB drives plus an OS drive. My needs continued to shrink and even this was more than I needed.
By far most of my drive space was used by video files. These, along with files being archived, were all that was on my Windows Home Server. All my non-video data had been moved to my Synology NAS.
The growth of streaming and cloud services meant my local video library rarely grew. Even in the rare cases where I bought a video, all else being equal, I’d prefer a cloud purchase and not have to worry about local storage. My Blu-Ray purchases for the past year could be counted on one hand.
I rarely accessed the WHS files, yet the server was running 24 hours a day, seven days a week. So as I was looking to downsize and save electricity, this was an obvious first choice.
So I cleaned up the files on my Synology DS1511+ NAS which I uses for backups and files storage and copied my video library to the available space. I had so many duplicate files and backups I was also able to free up another five 3 TB drives that were in an expansion unit and still have room for the WHS files.
So I copied the Windows Home Server files to the Synology 1511+ and then copied them to a few of the freed up drives to be put in storage as a backup. The Synology 1511+ just gets fired up every weekend to refresh backups and verify the drives still spin.
I moved a couple of the 3 TB drives to my Synology 212+ NAS which serves as my main data storage for what I consider my active data. The extra space will be used for time machine backups and future needs.
Windows Home Server will be supported into 2016 so there was no rush for me to replace it. Despite this, time has moved on and now my Synology NAS is better suited to my needs which doesn’t include needing terabytes of files being always available.
I’ve been testing an Airport Extreme Router and found that I couldn’t disable using it as a DHCP server. Well, actually if it’s in bridge mode it won’t be a DHCP server but I wanted it as a router, not a bridge.
I’m using what are the current Airport Extreme and Airport Utility. The Airport Utility version is 6.3.2 (632.3). The Airport Extreme shows a version of 7.7.3 (which I assume is the firmware but it just says version) and the part number is ME918LL/A. It is the current model at the time of this post (July 2014).
The three router modes are:
I wanted to use it as a NAT router but without DHCP. I already have a DHCP server I want to keep (my Synology NAS). I couldn’t really turn off DHCP but there was an easy enough way to get around this limitiation.
The short instructions are summarized in the screenshot below (details later).
1. Set a small DHCP range using IP addresses that aren’t used by the real DHCP server or any other device on the network. (Actually, it can probably duplicate another device but this is cleaner.) In the screenshot I used 192.168.1.253 to 192.168.1.254. I had to use two addresses, the same beginning and ending addresses weren’t allowed.
2. Create dummy DHCP reservations for each of the IP addresses. The MAC addresses don’t have to be real.
Open the Airport Extreme Utility and go to the Network tab and click the Network Options button.
Set a DHCP range that’s appropriate for your network. Use addresses that aren’t used by any of your computers or other DHCP ranges. (In theory none of these addresses should be used, but keeping things valid will avoid problems.) Save the screen and you’ll be back on the network tab.
Click the “+” sign under “DHCP Reservations.
Type in a description, make sure “MAC Address” is selected for “Reserve Address By” and type an dummy Mac address. I just type the number “1” (or 2) until it stops me. Then save the information.
Repeat step 4 for all IPs in the DHCP range. The Airport Utility will prefill an unreserved IP in the range so you don’t need to keep track.
Save everything all the way out and your Airport Extreme will restart.
In the end the Airport Extreme is still running a DHCP server, except it doesn’t have any IP addresses to hand out so the “real” DHCP will be the only one to respond.
I was doing my yearly backup review and decided to take a look at Backblaze. Backblaze is popular among Mac users although they also have a Windows client. I took a look at the Mac version.
Backblaze – The Basics
Backblaze is cloud backup. They offer “unlimited” backup but there are restrictions if not limitations. Pricing is $5 per month but is discounted for 1 year ($50) and 2 years ($95) prepaid commitments. Pricing is per computer.
Only locally attached disks can be backed up. This obviously includes internal drives and also includes firewire, USB and Thunderbolt attached drives. Network attached drives are not backed up.
There is a long list of folders on my Mac that cannot be backed up, even if I wanted to. The important ones in this list are “Applications” and “Library” (the system-wide library folder). So Backblaze cannot be used as a full system backup. On the positive side, everything that would be specific to a user can be backed up.
There’s also a long list of file types that are not backed by default although these exclusions can be removed:
On the Mac you may want to be sure sparseimage files are backed up depending on how you use them. I kept virtual machine images excluded because I didn’t want to use the bandwidth to backup constantly changing files. I did remove the ISO exclusion since any ISOs I have don’t change once they’re backed up.
Backblaze will delete files from cloud storage 30 days after they’re deleted from the computer. If a external drive is disconnected from the computer (or turned off) the files will be deleted from the backup after 30 days. I wasn’t able to test this during the trial period but it appears there’s now a warning message a couple weeks before the files from the external drive are deleted.
In a true “trust no one” philosophy I can set my own encryption key. (I do have to trust that the Backblaze software is not secretly sending them my encryption key.) If I lose this key I lose all ability to restore the files and Backblaze can’t get them back. Despite this risk I require this for all my cloud storage that contains my financial and confidential records and keep the key in multiple locations.
I took advantage of their two week trial to take a look at the Backblaze software and service. The registration, download and installation was straightforward so I won’t repeat it here. Once installed it began backing up using the default settings.
Backblaze installs itself as a preference pane. The options are (click any image for full size)…
The Main Panel
The main panel gives you an overview of where you stand with backups. By default Backblaze will continuously backup changed files. Clicking the settings button will display six configuration screens.
The computers online name is automatically generated and is based on the Mac’s name. I have only one drive so it’s the temporary data drive. I changed the warning from the default 7 days to 1 day. If it’s failing I want to know right away.
Any attached hard drives are listed. Check the ones you want included, uncheck the ones that you don’t.
Performance lets you throttle the bandwidth used during the backup. I found the Automatic setting worked well. The estimates varied widely during my first backup so the estimate seems to be based on a narrow slice of time. I had no problem streaming Amazon or Netflix while the backup was uploading. This isn’t too surprising since the data is going in opposite directions. The uploads did slow down during the streaming.
Nothing much here. I used “Continuously.”
The folders can’t be removed from the exclusion list but you can add other folders to exclude. I left the folders as is but I did remove DMG, ISO and SparseImage files from the file type exclusions.
This panel is primarily informational but it’s also where I entered my private encryption key.
There’s three report panels that are informational.
The Reports panel showing a summary of the files you have backed up.
Any files that are still waiting to be backed up are shown here.
Any backup problems are shown here
The restore panel is informational and shows the three restore options.
Web Restore Test
Naturally, a backup is no good if you can’t restore the files. So I went through the process of a web restore. It’s a straight-forward process, although a web restore may not be well suited to restoring gigabytes of data. I only picked a couple files for the restore and they were ready in a matter of minutes.
You do have to enter your encryption key on the restore web page. So you have to again trust that Backblaze won’t remember your encryption key. You also need to trust that there’s no coding errors that will compromise the key. This is probably an acceptable risk for most people. It also seems that the files are decrypted on the server side before they are put into the zip file. It’s also seems that if you request that the files be sent to you on disk that they’ll be sent already unencrypted on the disk which seems like a bigger risk.
The web restore process is shows in the following screenshots.
Logon to your web account and go to the restore option.
You’ll first pick the files that you want to restore.
You can select just a few files or all the files.
After selecting the files for restoration. You wait until the files are ready for restore.
You receive an email when the files are ready for restoration.
The files are also available on your account once they are ready. You download the zip file from here.
Backblaze is a nice easy to use Backup service. It shares similar limitations to other “all you can eat for one price” backup services. I can see why it’s so popular but I’d prefer a backup service that can be used for long term storage without having to worry about whether or not the drive is still connected. So this became a non-starter for me.
The service is designed to backup data. I didn’t try restoring any complete OS X package files. For example, a Bento database. In the case of iPhoto and Aperture libraries (which are also library files) Backblaze will look into the library and backup the photos. But it won’t backup thumbnails. While not backing up thumbnails isn’t a problem, this less than everything approach does concern me. I’m more comfortable when I know exactly what’s happening with my backups.
The low price of Backblaze backup and ease of use will appeal to most people and for good reason.
Easy to Use
Fast backup (no throttling noticed – limitation seemed to be my connection)
Web restore cumbersome for large restores
Large restores (beyond web capabilities) costly and delayed by shipping.
Private encryption key has some potential holes (required before restore so files are shipped unencrypted or stored unencrypted on the Backblaze server)
Pro/Con (depends on your point of view)
Like all cloud backups, potentially limited by your internet connection for both speeds and data caps.
Not a full disk backup, still need a second backup solution for a full system.
Deletes backed up files 30 days after the local files vanish.
I hate having software running when it’s not needed. I also hate always having to manually start and stop software based on some often repeated action.
I recently bought a ScanSnap scanner and the ScanSnap manager needs to run whenever I want to scan. Using Keyboard Maestro’s ability to react to USB devices being connected or disconnected I was able to manage this automatically rather than leaving the software always running, or having to manually run it. What’s cool is that the ScanSnap S1300i turns on when the cover is opened (and shuts off when it’s closed) and Keyboard Maestro sees the on/off as the connection (or disconnection) based on the cover opening and closing. It’s not the actual cable connection being detected.
Hint: Activate the USB device when setting up the trigger and it will put the name of the newly connected device into the name field.