As previously described, my Xperia X10i sometimes refuses to reconnect to WIND Home after trips outside of the WIND zone. Seems the phone gets confused as to which network to try.
The 'trick' of recycling the SIM card through another phone - GSM or AWS - does work. I've managed to fix my card a few times that way. Today I stumbled on a better way that doesn't involve a second handset.
1. Go to launcher and start "SIM Toolkit".
2. My version has "English", "French", and "Roaming Setting". Select "Roaming".
3. Tap "Select Visited Network".
4. Reselect "Roaming Setting".
5. Tap "Automatic".
The Xperia should re-acquire WIND Home at this point.
Note your handset may be sluggish for a minute, as it "catches up" to the changes that happened while it was offline.
Hopefully this saves somebody some grief.
Monday, February 28, 2011
Tuesday, February 15, 2011
Are all Android phones flawed?
After hours of research, I decided to get a Samsung Vibrant (T-Mobile version) for my wife to work on WIND.
Shortly after, I found the Vibrant can have an issue where it doesn't pick up WIND Home properly. You have to manually lock it to WIND Home, which disables WIND Away i.e. roaming.
My wife won't know how to do this, so this is a big problem.
Bell has the Vibrant. I calculated out WIND vs. Bell, and it's around $200-$300 higher at Bell depending on how long you go (1, 2 or 3 years). Not overwhelming, especially if it fixes the network problems and gives her transparent roaming when she does the occasional travelling.
But here's the kicker - the Bell version has problems too! Many Vibrant owners have complained that to keep the phone from randomly locking up several times a day, you have to lock the phone to 850 MHz only. Sounds kinda familiar, doesn't it?
Maybe Froyo fixes this, I don't know yet. Samsung took forever to push out Froyo and managed to botch their first attempt anyway, leaving few people able to test the fixes.
But what is the deal? Seriously - are all smartphones significantly flawed in some way these days? Screen problems, keyboard registration issues, network issues - is there a single Android phone out there that has not shipped with some sort of defect, limitation or flaw?
I can't remember too many things my Palm didn't do right - the occasional crash from third-party software, and one or two instances where a sync went bad and I had to manually re-categorize a bunch of memos. (I was on a plane anyway for one of them, not bad timing overall.) Hell, near the end of it's working life it even got into an endless reboot loop, again from third-party software.
But it always made calls, always connected, always got email, never (or hardly ever) froze or hung, worked with headsets fine, etc. Everything could be corrected with on-phone resets and a restore from BackupBuddy, which never failed.
Not so with Android, which lacks a solid, comprehensive backup utility. My Backup Pro is as close as I've seen so far, and it fails to restore about half the stuff correctly.
[Edit]: After consulting with WIND technical support, they confirmed the Vibrant still has the Home/Away bug. Locking the network works, but requires manual switching between Home and Away. Not great for my wife, but neither is having the phone freeze up on her five times daily.
They also explained that when using a non-WIND phone, roaming will likely not work properly. The WIND phones have settings that allow them to correctly select between WIND, Rogers, and T-Mobile (their USA roaming partner), but compatible non-WIND phones do not.
So if you take the occasional trip to the USA, you may have to manually set your phone to T-Mobile. Not hard for some, but well beyond the capabilities of many users. And something that is likely to affect all non-WIND-branded phones, not just the Vibrant.
[Edit 4-2012: My non-WIND Xperia X10i picks up T-Mobile in the USA just fine. No manual interventions required, it just works (like it should).]
Personally I'm worried about the rare situation where my wife is on a trip - road or plane - gets into trouble and her phone doesn't work because the settings are manually locked. These days, we rely on having these things more than people admit. Unless Bell has a fix for the freezing issue, I might not be buying a Vibrant after all.
[Additional]: xda-developers says that Froyo has fixed the freezing problem. I guess we'll see...
Shortly after, I found the Vibrant can have an issue where it doesn't pick up WIND Home properly. You have to manually lock it to WIND Home, which disables WIND Away i.e. roaming.
My wife won't know how to do this, so this is a big problem.
Bell has the Vibrant. I calculated out WIND vs. Bell, and it's around $200-$300 higher at Bell depending on how long you go (1, 2 or 3 years). Not overwhelming, especially if it fixes the network problems and gives her transparent roaming when she does the occasional travelling.
But here's the kicker - the Bell version has problems too! Many Vibrant owners have complained that to keep the phone from randomly locking up several times a day, you have to lock the phone to 850 MHz only. Sounds kinda familiar, doesn't it?
Maybe Froyo fixes this, I don't know yet. Samsung took forever to push out Froyo and managed to botch their first attempt anyway, leaving few people able to test the fixes.
But what is the deal? Seriously - are all smartphones significantly flawed in some way these days? Screen problems, keyboard registration issues, network issues - is there a single Android phone out there that has not shipped with some sort of defect, limitation or flaw?
I can't remember too many things my Palm didn't do right - the occasional crash from third-party software, and one or two instances where a sync went bad and I had to manually re-categorize a bunch of memos. (I was on a plane anyway for one of them, not bad timing overall.) Hell, near the end of it's working life it even got into an endless reboot loop, again from third-party software.
But it always made calls, always connected, always got email, never (or hardly ever) froze or hung, worked with headsets fine, etc. Everything could be corrected with on-phone resets and a restore from BackupBuddy, which never failed.
Not so with Android, which lacks a solid, comprehensive backup utility. My Backup Pro is as close as I've seen so far, and it fails to restore about half the stuff correctly.
[Edit]: After consulting with WIND technical support, they confirmed the Vibrant still has the Home/Away bug. Locking the network works, but requires manual switching between Home and Away. Not great for my wife, but neither is having the phone freeze up on her five times daily.
They also explained that when using a non-WIND phone, roaming will likely not work properly. The WIND phones have settings that allow them to correctly select between WIND, Rogers, and T-Mobile (their USA roaming partner), but compatible non-WIND phones do not.
So if you take the occasional trip to the USA, you may have to manually set your phone to T-Mobile. Not hard for some, but well beyond the capabilities of many users. And something that is likely to affect all non-WIND-branded phones, not just the Vibrant.
[Edit 4-2012: My non-WIND Xperia X10i picks up T-Mobile in the USA just fine. No manual interventions required, it just works (like it should).]
Personally I'm worried about the rare situation where my wife is on a trip - road or plane - gets into trouble and her phone doesn't work because the settings are manually locked. These days, we rely on having these things more than people admit. Unless Bell has a fix for the freezing issue, I might not be buying a Vibrant after all.
[Additional]: xda-developers says that Froyo has fixed the freezing problem. I guess we'll see...
Monday, February 14, 2011
Xperia X10i still loses WIND
Yes, my Xperia X10i still gets into a state where it refuses to connect to WIND after it loses the network.
This is incredibly annoying. I don't travel enough to remember to shut off the phone when going outside the city. Yet if I don't, I lose my connection until I get home and cycle my SIM through my old phone.
Sony will never fix it. They're releasing the Arc, which spells the end for X10 support, and have already said they're leaving Xperia phones stuck on Android 2.1 with no possibility for upgrades. Certainly not upgrades to fix 'minor' bugs to fringe AWS-enabled models to correct problems with a few country-specific regional networks.
But you know what? I stopped caring once I found out about the HTC Pyramid and Samsung Galaxy S II. These are the Droid X-like Android monsters I have always wanted but could never get. And they're both AWS-enabled for T-mobile.
Mockups have already told me that I can live with a 4.3" phone, but a 5.0" semi-tablet like the Dell Streak is just too big. And I've lusted after the Droid X since before it hit the shelves, but Verizon kept it all to themselves. (The HTC Evo was nice, but the Droid smoked it in side-by-side tests, and looked more badass to boot - although their Terminator-esque theme was admittedly a bit too much.)
Samsung, on the other hand, has usually released variants of it's latest phones for all carriers. My wife is probably getting a Samsung Vibrant - also known as the Galaxy S and Fascinate - for her birthday, although she doesn't know it. It's T-mobile enabled, and not without it's teething problems, but is just as nice as my X10i - and cheaper, now, thanks to advances in the market.
So why not get a Galaxy S II for myself, and start enjoying not only 4.3 beautiful inches of Super AMOLED goodness with bonus awe-inspiring dual-core processor, but a noticeable lack of gotta-have-two-screen-locks stupidity, the non-resetting-phonebook-name-search dumbness, the I'll-turn-on-anytime-I-feel-like-it display, Bluetooth headset scrambled-voice-gobbledygook, the sluggish I-can't-seem-to-lock-focus and takes-forever-to-take-a-picture camera, the insert-the-charged-battery-and-claim-it's-dead quirk, the I've-got-signal-but-can't-make-a-call game, and I-see-the-network-but-don't-feel-like-connecting-anymore bug? I'm a loyal guy but I don't see the value here.
Sory, SE. You had a shot at it, and screwed it up royally. There are too many other options out there to bother with you any more. Let's hope I can sell my Xperia on Kijiji for a decent price after the dual-cores arrive.
This is incredibly annoying. I don't travel enough to remember to shut off the phone when going outside the city. Yet if I don't, I lose my connection until I get home and cycle my SIM through my old phone.
Sony will never fix it. They're releasing the Arc, which spells the end for X10 support, and have already said they're leaving Xperia phones stuck on Android 2.1 with no possibility for upgrades. Certainly not upgrades to fix 'minor' bugs to fringe AWS-enabled models to correct problems with a few country-specific regional networks.
But you know what? I stopped caring once I found out about the HTC Pyramid and Samsung Galaxy S II. These are the Droid X-like Android monsters I have always wanted but could never get. And they're both AWS-enabled for T-mobile.
Mockups have already told me that I can live with a 4.3" phone, but a 5.0" semi-tablet like the Dell Streak is just too big. And I've lusted after the Droid X since before it hit the shelves, but Verizon kept it all to themselves. (The HTC Evo was nice, but the Droid smoked it in side-by-side tests, and looked more badass to boot - although their Terminator-esque theme was admittedly a bit too much.)
Samsung, on the other hand, has usually released variants of it's latest phones for all carriers. My wife is probably getting a Samsung Vibrant - also known as the Galaxy S and Fascinate - for her birthday, although she doesn't know it. It's T-mobile enabled, and not without it's teething problems, but is just as nice as my X10i - and cheaper, now, thanks to advances in the market.
So why not get a Galaxy S II for myself, and start enjoying not only 4.3 beautiful inches of Super AMOLED goodness with bonus awe-inspiring dual-core processor, but a noticeable lack of gotta-have-two-screen-locks stupidity, the non-resetting-phonebook-name-search dumbness, the I'll-turn-on-anytime-I-feel-like-it display, Bluetooth headset scrambled-voice-gobbledygook, the sluggish I-can't-seem-to-lock-focus and takes-forever-to-take-a-picture camera, the insert-the-charged-battery-and-claim-it's-dead quirk, the I've-got-signal-but-can't-make-a-call game, and I-see-the-network-but-don't-feel-like-connecting-anymore bug? I'm a loyal guy but I don't see the value here.
Sory, SE. You had a shot at it, and screwed it up royally. There are too many other options out there to bother with you any more. Let's hope I can sell my Xperia on Kijiji for a decent price after the dual-cores arrive.
Kudos to 3ware and the 9650SE
Although I suspected my storage server power supply wasn't up to the task, I nevertheless went ahead and loaded up the last 3 drives for a total of thirteen live drives, two hot spares and one cold spare.
The installation went really well, so I started the migration process and was about to stop worrying about it. However, during the migration a drive went and got some power problems, and ended up disconnecting both itself and a second drive from the array.
After several hundred error messages, I finally got down and removed the bad drive, but the damage was done - I now had a RAID6 volume in the middle of a migrate operation with two drives missing. The 3ware manual is silent on what happens in this case - it only says the migration process may not be aborted.
Kudos to the 3ware engineers, though. The 9650SE not only went ahead and completed the migration on the degraded array, it then immediately picked up the two hot spares and began a full rebuild. Hopefully no more drives will go down in the next 36 hours, which should be enough time to finish rebuilding. (Hey, it's a nine terabyte array, it takes a while.)
I have no doubt that if I had a third failure during the migrate, or have one during the rebuild, my array will be gone. That's a limitation of RAID, though, and not the controller.
This isn't the first time the RAID6 controller has saved my ass. Thanks to Seagate's ineptitude and Intel's inadequate understanding, I built my entire original array from ST34000340AS 1Tb 7200.11 drives that were all potential bricks. In the process of updating them all, I booted the server with a new 'fixed' drive, only to have a second drive brick on boot. This left me with two degraded drives, which would have taken my old RAID5 down. Fortunately, the RAID6 kept working and got it all sorted out.
A word to the wise, though - when rebuilding, set the controller to do it quickly. If I had set background tasks to a higher priority, it would have completed before the drive errors caused any issues.
So far very few of the original Seagate drives are left in my array - three of the eight, to be exact. They all failed before five years, meaning that two-thirds of them did not even make their warranty period. Hardly an impressive statistic, even leaving aside the firmware problems that caused so much grief.
Returning them to Seagate is uneconomic, and incredibly burdensome, so I just end up replacing them and buying the over-the-counter warranty at Memory Express to avoid future issues. Also mixing up the array with different brands and models, to avoid unnecessary exposure to future bugs.
Despite several attempts, fate has yet to permanently disable my server, thanks in no small part to the features and engineering of the 9650SE. Let's hope that 3ware is able to continue this tradition of fine products under their new ownership.
I do still think that my PSU is not quite up to the load, though. At the time, getting a redundant unit seemed more important than the overall wattage, and I think MemX did their best, but if they had any inkling that the box wasn't going to support my planned 16 drives they kept it to themselves. This puts me between a rock and a hard place on my next build - do I go for high output and get rock-solid hot-swap for more failed drives - but risk the entire box dying one day - or do I stay with redundant and take the risk of power-line sags affecting the array?
A PSU failure on the server can probably be fixed in four hours, and with adequate backups my entire array brought back within a day or two. But it really corrupts the original vision of a bulletproof, always-on server. Of course, not being able to effectively hot-swap drives and always worrying about dropouts due to poor power isn't comforting either.
I should probably go rack-mount with high wattage redundant supplies, but I don't want to make that kind of investment yet. I guess I'll stick with what I have and see where that leads.
The installation went really well, so I started the migration process and was about to stop worrying about it. However, during the migration a drive went and got some power problems, and ended up disconnecting both itself and a second drive from the array.
After several hundred error messages, I finally got down and removed the bad drive, but the damage was done - I now had a RAID6 volume in the middle of a migrate operation with two drives missing. The 3ware manual is silent on what happens in this case - it only says the migration process may not be aborted.
Kudos to the 3ware engineers, though. The 9650SE not only went ahead and completed the migration on the degraded array, it then immediately picked up the two hot spares and began a full rebuild. Hopefully no more drives will go down in the next 36 hours, which should be enough time to finish rebuilding. (Hey, it's a nine terabyte array, it takes a while.)
I have no doubt that if I had a third failure during the migrate, or have one during the rebuild, my array will be gone. That's a limitation of RAID, though, and not the controller.
This isn't the first time the RAID6 controller has saved my ass. Thanks to Seagate's ineptitude and Intel's inadequate understanding, I built my entire original array from ST34000340AS 1Tb 7200.11 drives that were all potential bricks. In the process of updating them all, I booted the server with a new 'fixed' drive, only to have a second drive brick on boot. This left me with two degraded drives, which would have taken my old RAID5 down. Fortunately, the RAID6 kept working and got it all sorted out.
A word to the wise, though - when rebuilding, set the controller to do it quickly. If I had set background tasks to a higher priority, it would have completed before the drive errors caused any issues.
So far very few of the original Seagate drives are left in my array - three of the eight, to be exact. They all failed before five years, meaning that two-thirds of them did not even make their warranty period. Hardly an impressive statistic, even leaving aside the firmware problems that caused so much grief.
Returning them to Seagate is uneconomic, and incredibly burdensome, so I just end up replacing them and buying the over-the-counter warranty at Memory Express to avoid future issues. Also mixing up the array with different brands and models, to avoid unnecessary exposure to future bugs.
Despite several attempts, fate has yet to permanently disable my server, thanks in no small part to the features and engineering of the 9650SE. Let's hope that 3ware is able to continue this tradition of fine products under their new ownership.
I do still think that my PSU is not quite up to the load, though. At the time, getting a redundant unit seemed more important than the overall wattage, and I think MemX did their best, but if they had any inkling that the box wasn't going to support my planned 16 drives they kept it to themselves. This puts me between a rock and a hard place on my next build - do I go for high output and get rock-solid hot-swap for more failed drives - but risk the entire box dying one day - or do I stay with redundant and take the risk of power-line sags affecting the array?
A PSU failure on the server can probably be fixed in four hours, and with adequate backups my entire array brought back within a day or two. But it really corrupts the original vision of a bulletproof, always-on server. Of course, not being able to effectively hot-swap drives and always worrying about dropouts due to poor power isn't comforting either.
I should probably go rack-mount with high wattage redundant supplies, but I don't want to make that kind of investment yet. I guess I'll stick with what I have and see where that leads.
Thursday, February 10, 2011
Brief review of the Moxa UPort 407 industrial 7-port USB hub
I've had the Moxa Uport 407 for a while now. Not sure why I didn't write about it before, guess it slipped my mind.
I got it because I needed a real powered USB hub. Not these rinky-dink pieces of crap you see everywhere from Best Buy to Memory Express, that purport to deliver 500 mA per port out of a 300 mA AC adapter.
(Seriously, what is it with USB and powered hubs? Do people really have to put up with that crap? Or do people basically not care? Probably the latter, as 99% of the people have no problems and don't know any better.)
Anyway, I cared because my USB peripherals were not behaving properly. I'd lose communication with my CD printer (A fantastic little Primera Z1, by the way, too bad they're discontinued), or my scanner, or whatever - randomly. The final straw was getting a Thermaltake BlacX USB drive dock that kind of crashed the whole thing.
So, after much reading, I went with the Moxa. Yes, they're expensive. And boy, was it worth it.
This thing is built like a tank, and comes with a massive AC adapter. Moxa says it can deliver the USB-specified 500 mA to every port - it can't, quite, since the AC adapter is a few mA short, but it's close enough for 99.9999% of the applications out there. The metal case is nice, and although it doesn't seem to get warm, it sure beats the stupid $14.99 USB hub my buddy had with the homebrew heatsink sticking out a hole hacked in the top.
Plugged it in, and hey! My CD printer is suddenly cycling itself on boot, it never did that before. And look! It works! No lost communications, no weird hiccups, no nothing, everything just works. Bliss!
They make a 4 port - personally, I needed the 7 port. I have all 7 ports full of whatever, and could probably use another 7. Everything is USB these days.
I've had it for over a year and it's been flawless. No signs of overheating, no hiccups, no problems of ANY kind. In fact, it's been so good I forgot it was there, which is kinda the whole point. I mounted it underneath my desk. The screw-lock on the power cable is a nice touch, as I never have to worry about kicking the power cord out by freak accident.
When I upgrade, if Moxa has a USB 3.0 solution I'll buy one in two seconds flat. Maybe two, so I can have 14 USB ports available. These things are what all USB hubs should be - bulletproof. Well worth it, IMHO.
[Edit]: Just asked Moxa, and they don't have any USB 3.0 roadmap yet. Damn.
I got it because I needed a real powered USB hub. Not these rinky-dink pieces of crap you see everywhere from Best Buy to Memory Express, that purport to deliver 500 mA per port out of a 300 mA AC adapter.
(Seriously, what is it with USB and powered hubs? Do people really have to put up with that crap? Or do people basically not care? Probably the latter, as 99% of the people have no problems and don't know any better.)
Anyway, I cared because my USB peripherals were not behaving properly. I'd lose communication with my CD printer (A fantastic little Primera Z1, by the way, too bad they're discontinued), or my scanner, or whatever - randomly. The final straw was getting a Thermaltake BlacX USB drive dock that kind of crashed the whole thing.
So, after much reading, I went with the Moxa. Yes, they're expensive. And boy, was it worth it.
This thing is built like a tank, and comes with a massive AC adapter. Moxa says it can deliver the USB-specified 500 mA to every port - it can't, quite, since the AC adapter is a few mA short, but it's close enough for 99.9999% of the applications out there. The metal case is nice, and although it doesn't seem to get warm, it sure beats the stupid $14.99 USB hub my buddy had with the homebrew heatsink sticking out a hole hacked in the top.
Plugged it in, and hey! My CD printer is suddenly cycling itself on boot, it never did that before. And look! It works! No lost communications, no weird hiccups, no nothing, everything just works. Bliss!
They make a 4 port - personally, I needed the 7 port. I have all 7 ports full of whatever, and could probably use another 7. Everything is USB these days.
I've had it for over a year and it's been flawless. No signs of overheating, no hiccups, no problems of ANY kind. In fact, it's been so good I forgot it was there, which is kinda the whole point. I mounted it underneath my desk. The screw-lock on the power cable is a nice touch, as I never have to worry about kicking the power cord out by freak accident.
When I upgrade, if Moxa has a USB 3.0 solution I'll buy one in two seconds flat. Maybe two, so I can have 14 USB ports available. These things are what all USB hubs should be - bulletproof. Well worth it, IMHO.
[Edit]: Just asked Moxa, and they don't have any USB 3.0 roadmap yet. Damn.
Brief review of the Icy Dock MB973SP-B 3-in-2 hot-swap trayless HDD bay
I got these for my home server. I knew I was taking a bit of a chance because of the mixed reviews, but I bought one module to try out.
These units are fine, IMHO. They do what they're supposed to do, and are well made. They're not crap is what I'm trying to say.
Yes, they could possibly be stronger, but what couldn't? If you're not stupid, and handle your stuff with reasonable care anyway, you will have zero problems with these units. They're not junk and work just fine, thank you very much.
You will want to try them out in your case first, if possible. They are large and need additional depth, plus the side panel design might not be compatible with some cases. In my case, the server uses drive rails, and these bays work fine.
I bought four of these, plus one 4-in-3 bay (MB674SPF-B). I have this giant, old doublewide server case that nobody has ever been able to identify, and it has four bays on the left and ten on the right, plus some internal bays for the system drives. Now it holds 16 external hard drives, plus a DVD drive, with room to spare.
Previously, it would take me at least an hour to down the system, take it apart (yes, it is semi-toolless) insert and attach the drives, put it back together and reboot. Well, I just stuck in the last four drives and it took me 5 minutes. That, to my mind, already paid for the hot swap bays. Not to mention being able to pull a failed drive some time ago in 1 minute.
Hot swapping has been a bit iffy, but I think that's a consequence of either the Escalade 9650SE-16 or the power supply not being up to the task. Some bays appear to hot-swap fine, but there is one individual bay in one 3-in-2 dock that gives my system fits. So you may need to test out if hot-swap is really going to work. Note this is more than likely NOT the fault of the bays, it's just a consequence of running a relatively large RAID array. You can always down the server for 5 minutes to upgrade.
I picked these units - and waited six months for them to become available - because they include a nice fan, which is easy to swap out, as well as a high-temp alarm. I have never heard the alarm, and I hope I never will, and I've never changed the fans, and I hope I never will. There's always a possibility that a fan will fail, but so far the only one I've actually seen was one on a CPU cooler, after over a decade of continuous use, so I'm not too worried.
I don't have SMART data on the drives because the 3ware doesn't give it to me - not sure why. But there appears to be enough airflow to keep everything cool. I previously had all the drives in iCage units, which are great, but don't expand the drive bay capacity. I can't tell any difference between the Icy Dock and the iCage exhaust air temp.
With that many fans it's going to be noisy. But if you have an 18-drive machine in your office, you expect that. Put it in the basement already. I don't even try to regulate the fan speeds, they're all set to high permanently.
I'd recommend these to anybody looking for something like this. As a bonus, they make a large system look pretty badass in an understated way, all professional, clean and symmetrical.
These units are fine, IMHO. They do what they're supposed to do, and are well made. They're not crap is what I'm trying to say.
Yes, they could possibly be stronger, but what couldn't? If you're not stupid, and handle your stuff with reasonable care anyway, you will have zero problems with these units. They're not junk and work just fine, thank you very much.
You will want to try them out in your case first, if possible. They are large and need additional depth, plus the side panel design might not be compatible with some cases. In my case, the server uses drive rails, and these bays work fine.
I bought four of these, plus one 4-in-3 bay (MB674SPF-B). I have this giant, old doublewide server case that nobody has ever been able to identify, and it has four bays on the left and ten on the right, plus some internal bays for the system drives. Now it holds 16 external hard drives, plus a DVD drive, with room to spare.
Previously, it would take me at least an hour to down the system, take it apart (yes, it is semi-toolless) insert and attach the drives, put it back together and reboot. Well, I just stuck in the last four drives and it took me 5 minutes. That, to my mind, already paid for the hot swap bays. Not to mention being able to pull a failed drive some time ago in 1 minute.
Hot swapping has been a bit iffy, but I think that's a consequence of either the Escalade 9650SE-16 or the power supply not being up to the task. Some bays appear to hot-swap fine, but there is one individual bay in one 3-in-2 dock that gives my system fits. So you may need to test out if hot-swap is really going to work. Note this is more than likely NOT the fault of the bays, it's just a consequence of running a relatively large RAID array. You can always down the server for 5 minutes to upgrade.
I picked these units - and waited six months for them to become available - because they include a nice fan, which is easy to swap out, as well as a high-temp alarm. I have never heard the alarm, and I hope I never will, and I've never changed the fans, and I hope I never will. There's always a possibility that a fan will fail, but so far the only one I've actually seen was one on a CPU cooler, after over a decade of continuous use, so I'm not too worried.
I don't have SMART data on the drives because the 3ware doesn't give it to me - not sure why. But there appears to be enough airflow to keep everything cool. I previously had all the drives in iCage units, which are great, but don't expand the drive bay capacity. I can't tell any difference between the Icy Dock and the iCage exhaust air temp.
With that many fans it's going to be noisy. But if you have an 18-drive machine in your office, you expect that. Put it in the basement already. I don't even try to regulate the fan speeds, they're all set to high permanently.
I'd recommend these to anybody looking for something like this. As a bonus, they make a large system look pretty badass in an understated way, all professional, clean and symmetrical.
Brief steps for Online Capacity Expansion on Escalade 9650SE-16 RAID controller
As there's not much in the way of information on this, here's a brief guide towards expanding your RAID array on the Escalade 9650 series controllers using 3ware Online Capacity Expansion and OpenSUSE 10.1.
Note that you can also use this to migrate between RAID5 and RAID6, assuming you have the correct number of extra drives.
0. Back up everything. [May be hard with big arrays - consider using a USB drive dock.]
1. Hook up your new drive(s). Usual disclaimers about not hot-swapping, etc, apply.
2. Log in to 3DM2 at http://192.168.1.111:888/
3. Go to Management/Maintenance.
4. Check your main RAID array and press "Migrate".
5. Check the drives you want to add. Select the correct RAID type (RAID5 or RAID6) and stripe size (64k stripe is default). Press OK.
6. Wait for migration. Can take a while.
7. Reboot to remount new RAID volume. The OS will not see the expansion yet.
8. Back up everything new.
9. Unmount your RAID array ('unmount' command).
9a. Start parted on your device (i.e. parted /dev/sdc), enter "check" as the command.
9b. Parted will complain the GPT table is not using all the available space. Hit "F" to fix it. This takes seconds.
10. Go to YaST2 Partition Manager.
11. Underneath the physical volume, select "linux native". Hit "Resize".
12. Set "Unused space" to zero. Apply.
13. Wait. (Maybe up to an hour - see below.)
14. Reboot.
Done! OpenSUSE should now see the expanded array size.
The steps are similar for Windows, although you will need to alter steps 9 through 12 to allow Windows to 'see' the expanded array.
Note that YaST Partitioner is a parted front-end, and parted ALWAYS resizes NON-destructively.
[Edit]: Well, I tried my own recipe, and it didn't work. It fails at step 12, where Yast Partitioner does not resize the partition so Opensuse can use the free space. I remember having this problem before, but it resolved itself, so the recipe above should be accurate. No dice this time, at least so far.
Trying GParted, I have found there is a GParted bug that prevents it from working. The fix is reportedly to run parted from the command line first. Run parted on the drive (i.e. /dev/sdc) and not the partition (i.e. /dev/sdc1). Solution thread is here.
[Edit]: Success! Steps 9a and 9b added above to fix the GPT table before trying Yast Partitioner. I'm sure GParted would work too, but it does basically what Yast Partitioner does.
For those wondering, the resize for a 10 Tb array took approximately 30 minutes. Horror stories about resize operations taking hours (or days) were, fortunately, not accurate in my case. This is also with the 3ware controller doing a background rebuild!
GParted uses e2fsck, so it may take considerably longer. Either that, or e2fsck takes a lot, lot less time than everybody says, even on multi-terabyte arrays.
Note that you can also use this to migrate between RAID5 and RAID6, assuming you have the correct number of extra drives.
0. Back up everything. [May be hard with big arrays - consider using a USB drive dock.]
1. Hook up your new drive(s). Usual disclaimers about not hot-swapping, etc, apply.
2. Log in to 3DM2 at http://192.168.1.111:888/
3. Go to Management/Maintenance.
4. Check your main RAID array and press "Migrate".
5. Check the drives you want to add. Select the correct RAID type (RAID5 or RAID6) and stripe size (64k stripe is default). Press OK.
6. Wait for migration. Can take a while.
7. Reboot to remount new RAID volume. The OS will not see the expansion yet.
8. Back up everything new.
9. Unmount your RAID array ('unmount' command).
9a. Start parted on your device (i.e. parted /dev/sdc), enter "check" as the command.
9b. Parted will complain the GPT table is not using all the available space. Hit "F" to fix it. This takes seconds.
10. Go to YaST2 Partition Manager.
11. Underneath the physical volume, select "linux native". Hit "Resize".
12. Set "Unused space" to zero. Apply.
13. Wait. (Maybe up to an hour - see below.)
14. Reboot.
Done! OpenSUSE should now see the expanded array size.
The steps are similar for Windows, although you will need to alter steps 9 through 12 to allow Windows to 'see' the expanded array.
Note that YaST Partitioner is a parted front-end, and parted ALWAYS resizes NON-destructively.
[Edit]: Well, I tried my own recipe, and it didn't work. It fails at step 12, where Yast Partitioner does not resize the partition so Opensuse can use the free space. I remember having this problem before, but it resolved itself, so the recipe above should be accurate. No dice this time, at least so far.
Trying GParted, I have found there is a GParted bug that prevents it from working. The fix is reportedly to run parted from the command line first. Run parted on the drive (i.e. /dev/sdc) and not the partition (i.e. /dev/sdc1). Solution thread is here.
[Edit]: Success! Steps 9a and 9b added above to fix the GPT table before trying Yast Partitioner. I'm sure GParted would work too, but it does basically what Yast Partitioner does.
For those wondering, the resize for a 10 Tb array took approximately 30 minutes. Horror stories about resize operations taking hours (or days) were, fortunately, not accurate in my case. This is also with the 3ware controller doing a background rebuild!
GParted uses e2fsck, so it may take considerably longer. Either that, or e2fsck takes a lot, lot less time than everybody says, even on multi-terabyte arrays.
Subscribe to:
Posts (Atom)