Us Warlocks can now give you back the advice you gave us at the start of WOTLK: if you don't like your class anymore, REROLL a different one and quit QQing.
I second this. It great for recovering data from pretty much anything: flash drives, memory cards, hard drives. The primary caveat is that the drive itself has to be fully operational. If the drive cannot be mounted/connected (like if the drive electronics are fried) this program won't help you.GetDataBack just does a scan of the disk and offers to 'undelete' any file fragments it finds.
Also, the bigger the drive, the longer it takes the process to complete. I suppose this is true of all tools that operate in this fashion so I can't say it's a con to GetDataBack.
Also, there's a separate version for FAT and for NTFS.
As an added bonus, if you have a Citrix server in a VM, disaster recovery is tons easier.
DR w/out VMWare:
1) Acquire new hardware 2) Build base OS 3) Service packs, reboot 4) Hotfixes, reboot 5) More hotfixes, reboot 6) etc, reboot 7) Use your build documentation to implement all the OS hacks needed to get that last bit of performance out of the box 8) Install Presentation Server 9) Citrix hotfixes, reboot 10) More Citrix hotfixes, reboot 11) Install your app 12) Add any tweaks/updates to the app 13) test and go into prod
DR with VMware: 1) Restore the vitual disk and VM config file 2) Build an ESX host server (~30 minutes) 3) Add the restored VM to the ESX host's console (if you're using virtual center its even easier) 4) Turn on the VM and you're done
* all of this assumes your DR location has all the necessary infrastructure like a directory service and your Citrix farm infrastructure is operational or has already been recovered
I just finished the ESX class. One of the labs was on VMotion, so I used that lab to try to "trick" VMotion into sucking.
My laptop ----> MS terminal svcs to VM1 My laptop ----> ping VM1
VM1 ----> ping VM2 VM1 ----> ping my laptop VM1 ----> terminal session to VM2
VM2 ----> ping VM1
When I did the Vmotion, I dropped one packet from my laptop to VM1. The other continuous pings didn't drop a single packet. The terminal session from my laptop to VM1 hitched for a half-second (I had grabbed a window w/ the mouse and was moving it around to see if I noticed any sort of delay).
Those 2 things were the sum total of noticeable effects of the Vmotion./me was impressed.
We have a similar situation - Novell file servers + Citrix servers + Netware client = 'teh suck'. In our case, we want to provide users access to their home drives, department drives, and a large mail archive storage cluster (all on Netware). As an aside, we're ditching Novell as fast as possible. Which isn't very fast. More like glacial. Anyway...
Dealing with the netware client was a pain and CSNW sucks for more than 1 server (users keep having to enter their -fairly long - context and tree) in an NDS tree as large as ours.
I proposed using a series of linux boxen to 'translate' Novell volumes into Samba shares that would be accessible without the Netware client. We didn't use the idea because we like pain and suffering. Maybe it will benefit you, though.
I would be VERY interested in the results of your testing w/ VMware and Citrix. We are about to embark on a similar mission...
If you are memory bound, adding more memory is the best way to increase capacity and/or performance.
Win2K3 Std 32-bit supports 4GB RAM, which is fine unless you need more than 4GB. Win2K3 Enterprise 32-bit supports like 32GB, but you'd have to pony up for the new OS (cost = cost of RAM + cost of Win2K3) Win2K3 Std 64-bit supports 6GB RAM (cost = cost of RAM)
Moving to 64-bit allows more memory usage without paying a larger M$ tax.
Citrix bought a company called Net6 a couple of years ago. Net6 made an SSL VPN "appliance", which runs a hardened Linux OS. Citrix rebranding it as the "Citrix Access Gateway", or CAG.
The 1st iteration was not so good because they rushed the rebranding and integration stuff. The 2nd and 3rd iterations were OK.
The latest revision is quite good. It supports around 2000 concurrent users, has easy to use yet powerful access controls and integrates nicely with Citrix's Presentation Server 4 product.
The cost is pretty good: the box is $2500 and licenses retail for around $100/concurrent user. If you have 100 users and your highest expected concurrent remote access count is 25, your cost would be $2500 + 25 x 100 = $5,000. If you buy 2 boxes (they have a built-in failover mechanism for redundancy), the cost would be $7500.
I work for a major healthcare provider and we're replacing Cisco VPN concentrators with the CAG. We bought 4 CAGs and are using Citrix's Advanced Access Control (AAC) product to integrate the CAGs with our internal portals (AAC makes the cost go up pretty high, though). We have around 40,000 users and our max concurrent remote users is currently around 4,000.
I'm a big proponent of using Citrix's Presentation Server as a means of providing secure access, any time, anywhere.
The basic idea is you build a Windows terminal server (more likely several - fault tolerance and all that), install Outlook on it, and use Citrix's web interface to provide a launch point. Outlook runs on the server, and any user with an ICA client (Citrix's client) can run it - even your Mac, *nix, and (gasp) DOS users.
That takes care of users inside the firewall. Outside, you have to have a VPN solution. Or, you can use the Citrix Access Gateway to provide access. By itself, it acts just like an SSL VPN, only cheaper (the box is $2500 and concurrent user licenses are $200 or so at retail). When used in conjunction with something called Advanced Access Control (AAC), you can provide secure access to any Citrix-delivered or web-delivered applications WITHOUT a VPN client. AAC turns the CAG into an HTTP/HTTPS/ICA proxy. It features RSA integration and all that jazz. As a side-bonus, AAC + CAG support smartphones, PDA's, and Crackberry's (though that last has an ugly UI).
Not the cheapest solution and it is certainly not as easy as I describe, but it is definitely capable of doing what you need - without having to retrain your users on a mail application. It is easier to say "go to this URL, log in, and click the Outlook icon" than it is to say, "here's PINE - Hope you remember how to use telnet":)
"Actually, each server should be able to handle the entire load of the cluster"
Good point. I took it as read that the first server in question is handling the load on its own and the question was solely about providing higher availability. I put in a note about clustered/load balanced servers being equivalent in available capacity because in my experience, this is a place that novices will shortsightedly cut corners to save $$$.
If you have a service that must be highly available, cluster or load balance the service. Use more than 1 box and either cluster them or load balance them.
RAID, ECC RAM, team NICs and all that stuff are very helpful, but if you want to make DARN sure that service is as available as possible, do server times two.
P.S. - your second server should be able to handle the exact same load as the first server or its not going to be terribly helpful
Slashdot Mirror
But not necessarily in that order.
It's only competitive if all players have access to such out-of-game training areas.
Us Warlocks can now give you back the advice you gave us at the start of WOTLK: if you don't like your class anymore, REROLL a different one and quit QQing.
Googling "recursion" always makes me laugh. On topic, I bet he means he has 2 hosts pinging each other and he Vmotion..err...Xenmotions one of them.
Or use a time machine to have the Bush administration fire off another "bailout".
Wouldn't it be more accurate to say that if no balls then yes it's a girl?
I second this. It great for recovering data from pretty much anything: flash drives, memory cards, hard drives. The primary caveat is that the drive itself has to be fully operational. If the drive cannot be mounted/connected (like if the drive electronics are fried) this program won't help you.GetDataBack just does a scan of the disk and offers to 'undelete' any file fragments it finds. Also, the bigger the drive, the longer it takes the process to complete. I suppose this is true of all tools that operate in this fashion so I can't say it's a con to GetDataBack. Also, there's a separate version for FAT and for NTFS.
I would pay for a browser that does this.
Agreed. This egregious display of laziness cannot be tolerated.
Only teenage geeks?
I officially feel old after reading your post.
Bean Burrito? I'd have KILLED for a bean burrito meal.
Ham slice was the new hotness back in my day. Now get off my lawn!
As an added bonus, if you have a Citrix server in a VM, disaster recovery is tons easier.
DR w/out VMWare:
1) Acquire new hardware
2) Build base OS
3) Service packs, reboot
4) Hotfixes, reboot
5) More hotfixes, reboot
6) etc, reboot
7) Use your build documentation to implement all the OS hacks needed to get that last bit of performance out of the box
8) Install Presentation Server
9) Citrix hotfixes, reboot
10) More Citrix hotfixes, reboot
11) Install your app
12) Add any tweaks/updates to the app
13) test and go into prod
DR with VMware:
1) Restore the vitual disk and VM config file
2) Build an ESX host server (~30 minutes)
3) Add the restored VM to the ESX host's console (if you're using virtual center its even easier)
4) Turn on the VM and you're done
* all of this assumes your DR location has all the necessary infrastructure like a directory service and your Citrix farm infrastructure is operational or has already been recovered
Well to be fair, the boxes were DL360's with only 4GB ram and another "team" was doing a Vmotion at the same time I did my testing.
There was 1 switch connecting all the servers together - an HP Procurve (and not a high end one, from what I understand)
I just finished the ESX class. One of the labs was on VMotion, so I used that lab to try to "trick" VMotion into sucking.
/me was impressed.
My laptop ----> MS terminal svcs to VM1
My laptop ----> ping VM1
VM1 ----> ping VM2
VM1 ----> ping my laptop
VM1 ----> terminal session to VM2
VM2 ----> ping VM1
When I did the Vmotion, I dropped one packet from my laptop to VM1. The other continuous pings didn't drop a single packet. The terminal session from my laptop to VM1 hitched for a half-second (I had grabbed a window w/ the mouse and was moving it around to see if I noticed any sort of delay).
Those 2 things were the sum total of noticeable effects of the Vmotion.
We have a similar situation - Novell file servers + Citrix servers + Netware client = 'teh suck'. In our case, we want to provide users access to their home drives, department drives, and a large mail archive storage cluster (all on Netware). As an aside, we're ditching Novell as fast as possible. Which isn't very fast. More like glacial. Anyway...
Dealing with the netware client was a pain and CSNW sucks for more than 1 server (users keep having to enter their -fairly long - context and tree) in an NDS tree as large as ours.
I proposed using a series of linux boxen to 'translate' Novell volumes into Samba shares that would be accessible without the Netware client. We didn't use the idea because we like pain and suffering. Maybe it will benefit you, though.
I would be VERY interested in the results of your testing w/ VMware and Citrix. We are about to embark on a similar mission...
If you are memory bound, adding more memory is the best way to increase capacity and/or performance.
Win2K3 Std 32-bit supports 4GB RAM, which is fine unless you need more than 4GB.
Win2K3 Enterprise 32-bit supports like 32GB, but you'd have to pony up for the new OS (cost = cost of RAM + cost of Win2K3)
Win2K3 Std 64-bit supports 6GB RAM (cost = cost of RAM)
Moving to 64-bit allows more memory usage without paying a larger M$ tax.
Test it, though. Then, test some more.
Redneck is a race?
Yes, Homo Sapiens NASCAR.
Yeah, they also dropped their price to 5 bucks under Citrx's price for the CAG so they can compete.
Looks like a good deal for the 50 user office, though I don't know any more about it than what I found with teh Google.
Citrix bought a company called Net6 a couple of years ago. Net6 made an SSL VPN "appliance", which runs a hardened Linux OS. Citrix rebranding it as the "Citrix Access Gateway", or CAG.
t .asp?contentID=15005
The 1st iteration was not so good because they rushed the rebranding and integration stuff. The 2nd and 3rd iterations were OK.
The latest revision is quite good. It supports around 2000 concurrent users, has easy to use yet powerful access controls and integrates nicely with Citrix's Presentation Server 4 product.
The cost is pretty good: the box is $2500 and licenses retail for around $100/concurrent user. If you have 100 users and your highest expected concurrent remote access count is 25, your cost would be $2500 + 25 x 100 = $5,000. If you buy 2 boxes (they have a built-in failover mechanism for redundancy), the cost would be $7500.
I work for a major healthcare provider and we're replacing Cisco VPN concentrators with the CAG. We bought 4 CAGs and are using Citrix's Advanced Access Control (AAC) product to integrate the CAGs with our internal portals (AAC makes the cost go up pretty high, though). We have around 40,000 users and our max concurrent remote users is currently around 4,000.
Check it out: http://www.citrix.com/English/ps2/products/produc
And no, I'm not the CEO of Citrix in disguise. I just believe in their products; we've saved a ton of $$$ using them!
I'm a big proponent of using Citrix's Presentation Server as a means of providing secure access, any time, anywhere.
:)
The basic idea is you build a Windows terminal server (more likely several - fault tolerance and all that), install Outlook on it, and use Citrix's web interface to provide a launch point. Outlook runs on the server, and any user with an ICA client (Citrix's client) can run it - even your Mac, *nix, and (gasp) DOS users.
That takes care of users inside the firewall. Outside, you have to have a VPN solution. Or, you can use the Citrix Access Gateway to provide access. By itself, it acts just like an SSL VPN, only cheaper (the box is $2500 and concurrent user licenses are $200 or so at retail). When used in conjunction with something called Advanced Access Control (AAC), you can provide secure access to any Citrix-delivered or web-delivered applications WITHOUT a VPN client. AAC turns the CAG into an HTTP/HTTPS/ICA proxy. It features RSA integration and all that jazz. As a side-bonus, AAC + CAG support smartphones, PDA's, and Crackberry's (though that last has an ugly UI).
Not the cheapest solution and it is certainly not as easy as I describe, but it is definitely capable of doing what you need - without having to retrain your users on a mail application. It is easier to say "go to this URL, log in, and click the Outlook icon" than it is to say, "here's PINE - Hope you remember how to use telnet"
To provide robust access anytime, anywhere
...Satan supporting the bible.
subject says it all. Guess I enjoy getting modded down.
"Actually, each server should be able to handle the entire load of the cluster"
Good point. I took it as read that the first server in question is handling the load on its own and the question was solely about providing higher availability. I put in a note about clustered/load balanced servers being equivalent in available capacity because in my experience, this is a place that novices will shortsightedly cut corners to save $$$.
If you have a service that must be highly available, cluster or load balance the service. Use more than 1 box and either cluster them or load balance them.
RAID, ECC RAM, team NICs and all that stuff are very helpful, but if you want to make DARN sure that service is as available as possible, do server times two.
P.S. - your second server should be able to handle the exact same load as the first server or its not going to be terribly helpful
You need to bone up on your Sci-Fi literature.