@Chopper3 plugging everything into a switch and powering it up is complex? The whole IP configuration / naming / target portal setup stuff for iSCSI comes on top of everything you would need to do for FC as well
@syneticondj that's really not how to do FC at all - if you throw everything into one big hub you'll run into all manner problems - you need VSANs, zones, zonesets, aliases - the lot for it to work properly - we have people that ONLY manage our fc switches, that's all they do - done right they're super complex
@Chopper3 this is for a complex configuration - with iSCSI it would not look vastly more elegant. The quite common "one target, three hosts" requirement would not need any of that, though - I often hear from people running FC because it "just works".
The barriers to entry for FC are higher, though. Special switches, cabling and HBAs compared to regular L2 gear for iSCSI that people are generally more familiar with
When I was at the Storage Networking World conference in Frankfurt at the advent of iSCSI every FC vendor I talked to was saying pretty much the same words to me: "iSCSI is new and complex, FC is established and much simpler to deploy"
And this apparently is the perception of most small-scale installation admins too
I can only speak having come at it from a Cisco perspective (I'm a CDCSNDS) and I have no idea at the small-scale end so we'll have to agree to disagree
@MDMarra people really familiar with Ethernet and IP would run iSCSI in a sleepwalk. But many admins do not even know about RMON counters, so it is kind of a stalemate - if you are mostly ignorant in both areas, you would choose the technology which is supposedly easier to manage
@Zoredache ATAoE was a really dumb idea. Some SMB appliances use it, but I gladly see it die.
Why is it a dumb idea? Theoretically it give you a very simple protocol, and possibly save you a few bytes since you get to skip all the IP encapsulation.
@ChrisS HyperSCSI has fallen into oblivion long ago. ATAoE probably is heading the same route. iSCSI was there. It was working all along. It had broad support amongst OSes and storage vendors alike. It had no significant shortcomings. Why the hell was there a need for another protocol doing the same thing but doing it worse?
They needed to to get any attention for it. The number of available devices is negligible though - there is Coraid of course and a handful of Chinese-made re-branded SOHO devices. At any rate, it is no match to what amount and range of stuff you get with iSCSI stacks nowadays
BTW: what about the survey? We really asked for it? (I am quite sure I did not)
And who is the target audience for the questionnaires? The regulars or the anonymous users?
I had a great – albeit brief – conversation with Matt Simmons while he was in town on his world tour in NYC recently that got me thinking about bringing the SysAdmin community together. This isn’t the first time that I’ve thought about how bring the SysAdmin community together in a stronger way. And, since [...]
It has no real value and just annoys the people. If it is just "for fun" then all it does is distract attention and burn productive time of millions of people visiting the site
@syneticondj It's the pop-up that's annoying, not the survey. That and the disrespect that you can't easily say "Your site is difficult to use because you keep INTRUSIVELY asking me how to make it better!"
@syneticondj What is "no real value"? Whose to say that some information can't be gleaned from it that may be useful? I think you're just getting ahead of yourself, and don't want to participate just for the sake of not participating.
@Holocryptic Statistics are difficult to get right. If you do not get them right, your conclusions will be based on wrong assumptions. If you are not able to tell if you got it right, you won't even know.
And it is not about me participating or not. I was just concerned about the annoyance potential - which is not as bad as I have initially assumed as it seems
BTW: it is by no means "my desire not to have a survey". It is my desire not to have an idiotic survey wasting everybody's time. I am quite fine with either a proper scientific survey or a survey people really choose to take without being molested beforehand.
@syneticondj & @MDMarra Cost is still an appropriate answer. The iSCSI spec is 257 pages long. The ATAoE spec is 12 pages from start to finish. ATA is the most widely support disk access protocol to date. ATAoE runs over standard Ethernet, nothing fancy. The fact that it can't be (natively) routed over the net is very comparable to Fibre Channel (which is wildly popular).
The main problem with ATAoE is the ATA protocol, it's shortcomings, especially for shared storage and the like.
FC, iSCSI, and HyperSCSI all have the advantages of SCSI... there's no getting around that. HyperSCSI offers no benefits over Fibre Channel besides cost. iSCSI has obvious advantages over HyperSCSI (the IP layer)... which is why FC and iSCSI are basically the only two out there today.
Speaking of the SCSI advantage, when are we going to ditch SATA and just go SAS for everything? SAS chips are basically only more expensive at this point because SATA has the economies of scale going for it.
had a slightly odd dream - I got a bunch of rats, killed them and put them in a big bag full of maggots, left them for a few weeks then tipped the lot into a blender already half full of dog shit then sprayed the remaining slurry through the letterbox of one of our neighbours - what's odd is that I quite like her, weird eh?
This is a Canonical Question about RAID levels, since there isn't a single definitive one.
What are:
the RAID levels typically used (including the RAID-Z family)?
deployments are they commonly found in?
benefits and pitfalls of each?
I have:
Openfiler SAN
ML370 G5
Smart Array 6400
slot 1
Array B
Logical Drive 2 is a RAID5 array of 6 x 148GB 10k hot plug drives making 680GB
no spare
Last Friday, the power went out, this machine was just plugged in to the wall and it went down hard. When it came back up, drive 1 and 4 of 0-5...
I have a HP Server with SmartArray P400 controller (incl. 256 MB Cache/Battery Backup) with a logicaldrive with replaced failed physicaldrive that does not rebuild.
This is how it looked when I detected the error:
~# /usr/sbin/hpacucli ctrl slot=0 show config
Smart Array P400 in Slot 0 (Embedd...
there was that guy who insisted that you must be able to restore a RAID 0 array by replacing the hard disk because he really needed to. That was a fun one.
Possible Duplicate:
What are the different widely used RAID levels and when should I consider them?
I have 6 SAS Harddrive configured on RAID 0. One of my SAS Harddrive has gone faulty, So the server is not working. I tried installing windows server 2003 on a SATA harddisk then connect...
I believe you can make this work, but it adds a bit of risk to your setup. You would need to add IIS to your domain controllers. Then redirect the index.asp to www.yourdomain.com. If you have multiple DCS, you'd need it on each.
I do not recommend this approach, but instead tell folks is cann...
I'm running into a maximum throughput issue here and need some advice on which way to tune my knobs. We're running a 10Gbit fileserver for backup distribution. It's a two disk S-ATA2 setup on an LSI MegaRAID Controller. The server also got 24gig of memory.
We have a need to mirror our last uploa...
The Hewlet-3Compaqard merge made infinite sense to me. Taking the power supply out of a HP PA-RISC machine nearly severed one guy's finger. *EVERY* Compaq device I've ever owned has caused me to bleed if I open the case and 3com... "ep0: Erase pencil mark!"
We've got a Macbook Air (Actually, we've got some Airs, Some Pros and some iMacs). We've got a consistently annoying problem, which is that when they connect to our DHCP server, they've actually got 2 different IP addresses, and that causes 2 different IP addresses to get set in our DHCP lease f...
Prague metro is really nasty. I've witnessed a drunk fall down the escalator stairs there once. He looked bad... But still was able to talk and refused any kind of help.
that poor guy is in a shit situation though - it sounds like his company is one of those "Give the developers whatever they ask for (even if it makes the environment unstable)" places
@voretaq7 Yep. I've been in that sort of situation before "This is our Star Employee, so he gets whatever he wants because his productivity is more important than the stability or longevity of our environment"
I have 3 different folders containing particular information about sales orders. Everything was working fine but a few days ago I started having an issue about the amount of subdirectories in each of those main folders (orders above 32K).
My temporary solution was to move the oldest data to a b...
This is a Canonical Question about RAID levels, since there isn't a single definitive one.
What are:
the RAID levels typically used (including the RAID-Z family)?
deployments are they commonly found in?
benefits and pitfalls of each?
I have a HP Server with SmartArray P400 controller (incl. 256 MB Cache/Battery Backup) with a logicaldrive with replaced failed physicaldrive that does not rebuild.
This is how it looked when I detected the error:
~# /usr/sbin/hpacucli ctrl slot=0 show config
Smart Array P400 in Slot 0 (Embedd...
Possible Duplicate:
What are the different widely used RAID levels and when should I consider them?
I have 6 SAS Harddrive configured on RAID 0. One of my SAS Harddrive has gone faulty, So the server is not working. I tried installing windows server 2003 on a SATA harddisk then connect...
I believe you can make this work, but it adds a bit of risk to your setup. You would need to add IIS to your domain controllers. Then redirect the index.asp to www.yourdomain.com. If you have multiple DCS, you'd need it on each.
I do not recommend this approach, but instead tell folks is cann...
Having a filter for no description would be helpful because then the tag editors can quickly tell which tags they need to make desciptions for.
I have 3 different folders containing particular information about sales orders. Everything was working fine but a few days ago I started having an issue about the amount of subdirectories in each of those main folders (orders above 32K).
My temporary solution was to move the oldest data to a b...
