I sure hope that Carbonite for Windows users have better luck than the version for the Mac. Carbonite definitely got crossed off my Christmas wishlist. After 4 days the initial backup has still not finished, and can't seem to figure out how much data was backed up or how much is left. My initial backup estimation was slightly more than 24 hours to back up the roughly 3GB of data. As of last night it said it was 85% complete, but now this afternoon it claims it has backed up 900MB with another 900MB to go.
Something stinks in Denmark.
Friday, December 19, 2008
Tuesday, December 16, 2008
Carbonite for Mac
Carbonite is now offering a free beta trial for Mac users.
Several years ago I tossed out my myriad of Windows, Linux, and "other" systems and switched to a Mac. MacOS X has been good to me, but solid Mac based "SOHO" quality backup applications have been seriously lacking. The best so far has been Time Machine on my new iMac but it still lacks easy set and forget offsite portability. I used to work for a "managed storage as a service" company that was entering the remote office / small office / home power user market using Avamar. I always thought that was a very cool solution but it lacked support for Mac's and wasn't without its bugs.
Carbonite is now offering a free public beta test of its Mac client, so I went and signed up. Signing up is easy, just basic details to get you logged in, and immediately presents you with a Mac based installer. Installing couldn't be easier. Just double click the package you just downloaded and off it goes. It doesn't even require you to re-enter any information, as the package apparently contains your registration information. The default is to automatically back up everything in your User folder. I chose to select what gets backed up, since I have a bunch of things I didn't want to have backed up, like that ISO image of Solaris 10 sitting on my desktop. Selections were easy enough to make, though the selection list didn't always update the icon on certain items after I unselected them even though the buttom to select them again did light up.
Overall my first impression is pretty good. The initial backup is expected to take at least 24 hours (hey, I have a lot of iTunes music and photos from my worldly travels!), afterwich I'll report back on how incremental backups and restores work. One thing to note, you can only select items under your personal Users folder on your primary drive. My external drive I use for Time Machine was selectable but my personal NAS sitting next to it was not, so I assume local USB and Firewire drives could be backed up without the same limitation as the internal OS drive.
Several years ago I tossed out my myriad of Windows, Linux, and "other" systems and switched to a Mac. MacOS X has been good to me, but solid Mac based "SOHO" quality backup applications have been seriously lacking. The best so far has been Time Machine on my new iMac but it still lacks easy set and forget offsite portability. I used to work for a "managed storage as a service" company that was entering the remote office / small office / home power user market using Avamar. I always thought that was a very cool solution but it lacked support for Mac's and wasn't without its bugs.
Carbonite is now offering a free public beta test of its Mac client, so I went and signed up. Signing up is easy, just basic details to get you logged in, and immediately presents you with a Mac based installer. Installing couldn't be easier. Just double click the package you just downloaded and off it goes. It doesn't even require you to re-enter any information, as the package apparently contains your registration information. The default is to automatically back up everything in your User folder. I chose to select what gets backed up, since I have a bunch of things I didn't want to have backed up, like that ISO image of Solaris 10 sitting on my desktop. Selections were easy enough to make, though the selection list didn't always update the icon on certain items after I unselected them even though the buttom to select them again did light up.
Overall my first impression is pretty good. The initial backup is expected to take at least 24 hours (hey, I have a lot of iTunes music and photos from my worldly travels!), afterwich I'll report back on how incremental backups and restores work. One thing to note, you can only select items under your personal Users folder on your primary drive. My external drive I use for Time Machine was selectable but my personal NAS sitting next to it was not, so I assume local USB and Firewire drives could be backed up without the same limitation as the internal OS drive.
Saturday, November 15, 2008
Netbackup Catalog Recovery
I left precise instructions, but they still managed to blow up the Netbackup catalog.
They were importing catalog images into a remote DR environment. It didn't go as planned, and it corrupted the NBDB instance of the ASA database. The image of someone slamming a square peg into a round hole with a hammer until it went through came into my head but I confidently went about my business and tried my best it get the system back online.
First of all, this is EXACTLY why I take a catalog backup of every Netbackup server in the environment. Second of all, I'm still surprised about the lack of documentation and lack of internal knowledge of Symantec support folks on how the new Netbackup Catalog introduced in 6.0 works.
I'm also surprised that they still haven't figured out Disaster Recovery. No, I don't mean how to recover the catalog in the event of a catastrophic loss like we suffered, but the fact of how we were using a home grown, and dangerous, script to import catalog images at our DR site instead of an official application extension. In this day and age, you would think that Netbackup would be able to syncronize certain clients or policies across multiple master servers for seamless recovery.
One can only hope.
They were importing catalog images into a remote DR environment. It didn't go as planned, and it corrupted the NBDB instance of the ASA database. The image of someone slamming a square peg into a round hole with a hammer until it went through came into my head but I confidently went about my business and tried my best it get the system back online.
First of all, this is EXACTLY why I take a catalog backup of every Netbackup server in the environment. Second of all, I'm still surprised about the lack of documentation and lack of internal knowledge of Symantec support folks on how the new Netbackup Catalog introduced in 6.0 works.
I'm also surprised that they still haven't figured out Disaster Recovery. No, I don't mean how to recover the catalog in the event of a catastrophic loss like we suffered, but the fact of how we were using a home grown, and dangerous, script to import catalog images at our DR site instead of an official application extension. In this day and age, you would think that Netbackup would be able to syncronize certain clients or policies across multiple master servers for seamless recovery.
One can only hope.
Monday, September 22, 2008
Disaster Recovery Planning
I had the pleasure of attending Jon Toigo's seminar on Disaster Recovery Planning put on by Techtarget / Storage Decisions when they rolled into Boston. I would definitely recommend you check it out next time they put one on near you. I've been reading his blog over the last year and found him quite entertaining in person.
Jon's presentation style is a bit confrontational when it comes to storage vendors, but valueable nonetheless. His emphasis on being able to roll your own solutions for just about everything was a bit offputting to a lot of people but I don't think the point was that you SHOULD roll your own on every solution but that you CAN, so think about where you want to spend the effort and where you want to spend your money. Small budget strapped shops need not spend a fortune to get the same value if you can spare the elbow grease.
In this dramatically sinking economy nobody wants to spend money they don't have to. The good news is with proper planning and evaluation you don't need to spend a fortune to create the safety net you need to keep your business from suffering a catastrophic loss during a timeframe when even once strong businesses are expected to be picked off one by one.
Jon's presentation style is a bit confrontational when it comes to storage vendors, but valueable nonetheless. His emphasis on being able to roll your own solutions for just about everything was a bit offputting to a lot of people but I don't think the point was that you SHOULD roll your own on every solution but that you CAN, so think about where you want to spend the effort and where you want to spend your money. Small budget strapped shops need not spend a fortune to get the same value if you can spare the elbow grease.
In this dramatically sinking economy nobody wants to spend money they don't have to. The good news is with proper planning and evaluation you don't need to spend a fortune to create the safety net you need to keep your business from suffering a catastrophic loss during a timeframe when even once strong businesses are expected to be picked off one by one.
Wednesday, July 9, 2008
Virtually Worthless
Having been evaluating a new platform for one of the major applications my group supports I've had the "pleasure" of sitting through numerous pitches by various vendors representing server, storage, and backup products. They have all been pushing their own version of virtualization. Virtualization has its place, but why does everything have to virtualized?
Virtual machines have been around for a long time. VMware certainly wasn't the first to do it, but they were quite innovative to do so on the x86 platform. But when you are talking hefty applications on big iron, introducing another layer of complexity doesn't solve my problem.
I don't need a gazillion little machines. I need a few that can support huge loads of I/O and all at once. I also need to not impact an application that runs the factory floor when I go to replace that failed NIC on a development web server. Don't get me wrong, squeezing more out of the hardware isn't a bad idea, I just don't like the idea of putting all my eggs in one basket. This is one thing that VMware managed to improve on by introducing load balancing and VM farms with ESX but it isn't without its caveats.
If I were to build a solution with all the virtulization techniques that all the vendors threw at us, I doubt any of it would work. Vendors need to do a better job at really figuring out how their virtualization options solve real world problems. None of the pitches I've heard convinced me they could fix my real world business or technical problems. Heck, I don't think any of them bothered to ask us about OUR problems before pushing virtualization.
Virtual machines have been around for a long time. VMware certainly wasn't the first to do it, but they were quite innovative to do so on the x86 platform. But when you are talking hefty applications on big iron, introducing another layer of complexity doesn't solve my problem.
I don't need a gazillion little machines. I need a few that can support huge loads of I/O and all at once. I also need to not impact an application that runs the factory floor when I go to replace that failed NIC on a development web server. Don't get me wrong, squeezing more out of the hardware isn't a bad idea, I just don't like the idea of putting all my eggs in one basket. This is one thing that VMware managed to improve on by introducing load balancing and VM farms with ESX but it isn't without its caveats.
If I were to build a solution with all the virtulization techniques that all the vendors threw at us, I doubt any of it would work. Vendors need to do a better job at really figuring out how their virtualization options solve real world problems. None of the pitches I've heard convinced me they could fix my real world business or technical problems. Heck, I don't think any of them bothered to ask us about OUR problems before pushing virtualization.
Sunday, July 6, 2008
No really, backups are cool?
I have been blogging in one form or another in one forum or another for a while now. Normally the jist of any work related content is something along the lines of why my job is driving me crazy, but I've decided to start to write some technical content in hopes that someone will find it useful or interesting in some way. I can't promise non-technical content won't rear its ugly head nor will I promise what I write will make Mr. Ruchinski, my high school English teacher who inspired me to start writing in the first place, proud of my grasp of the English language.
Of course you probably want to know that I'm not just some recent newcomer to the technology sector who thinks he knows everything. I've been thinking I know everything for about 14 years now. 16 if you count the "computer stuff" I did before officially entering the workforce. The longer I go the more I learn and the more I realize I didn't know as much as I thought I did. I've never had a job where I didn't learn something, even if the job sucked. I like to learn and am glad I can work in IT. Like a lot of others I've spent more than half of my career as either a contractor or a consultant for some sort of outsourcing/"managed services" provider. Moving from job to job gave me a pretty wide base of experience and like to think I could pick up any new environment without much of a learning curve. My true passion though is backups. Most people hate backups. It is a chore. Tedious and laborious. But with the right software, hardware, and know how, you can create a system that works well more often than not.
Now I suppose that can go for any "system" one can design, but backups are something I've personally figured out how to do particularly well. I did it the hard way. It is through blood sweat and tears that I have the knowledge that I do. On the first day of my current job, it took me about 3 hours to come up with a list of 14 things wrong with the environment. By my second week on the job I probably identified another 14. Seven years prior I wouldn't have noticed any of them.
No really, backups are cool? Well, yeah. Backups are cool because they can save someones ass. Backups let me feel comfortable with letting a junior analyst loose with administrative privileges. Backups give you the ability to become a hero. And they are a heck of a lot more complicated than most people realize, even people who are typically tasked with maintaining the backups. It is a process that touches and sometimes even stresses out every piece of your datacenter. You have to have a firm grasp of all the pieces to do it right and do it better.
Alas, the "Backups Are Cool" moniker didn't come from my own mind. That credit goes to an Engineer from Datadomain. I don't recall his name, but he got quite excited about talking about backup optimization until he realized I was the only person there not giving him a weird look. Seriously dude, backups ARE cool.
Of course you probably want to know that I'm not just some recent newcomer to the technology sector who thinks he knows everything. I've been thinking I know everything for about 14 years now. 16 if you count the "computer stuff" I did before officially entering the workforce. The longer I go the more I learn and the more I realize I didn't know as much as I thought I did. I've never had a job where I didn't learn something, even if the job sucked. I like to learn and am glad I can work in IT. Like a lot of others I've spent more than half of my career as either a contractor or a consultant for some sort of outsourcing/"managed services" provider. Moving from job to job gave me a pretty wide base of experience and like to think I could pick up any new environment without much of a learning curve. My true passion though is backups. Most people hate backups. It is a chore. Tedious and laborious. But with the right software, hardware, and know how, you can create a system that works well more often than not.
Now I suppose that can go for any "system" one can design, but backups are something I've personally figured out how to do particularly well. I did it the hard way. It is through blood sweat and tears that I have the knowledge that I do. On the first day of my current job, it took me about 3 hours to come up with a list of 14 things wrong with the environment. By my second week on the job I probably identified another 14. Seven years prior I wouldn't have noticed any of them.
No really, backups are cool? Well, yeah. Backups are cool because they can save someones ass. Backups let me feel comfortable with letting a junior analyst loose with administrative privileges. Backups give you the ability to become a hero. And they are a heck of a lot more complicated than most people realize, even people who are typically tasked with maintaining the backups. It is a process that touches and sometimes even stresses out every piece of your datacenter. You have to have a firm grasp of all the pieces to do it right and do it better.
Alas, the "Backups Are Cool" moniker didn't come from my own mind. That credit goes to an Engineer from Datadomain. I don't recall his name, but he got quite excited about talking about backup optimization until he realized I was the only person there not giving him a weird look. Seriously dude, backups ARE cool.
Subscribe to:
Posts (Atom)