Reputation
5,405
Top tag
Next privilege 10,000 Rep.
Access moderator tools
Badges
1 7 27
Newest
 Student
Impact
~394k people reached

Apr
28
comment Software vs hardware RAID performance and cache usage
You're imagining. You didn't give clear answer. There's no understanding in what you say, just "learnt-by-heart" what you've been told once. No thinking, just replaying what you heard from someone.
Apr
28
comment Software vs hardware RAID performance and cache usage
I asked you very clear and simple question. I don't see clear and simple answer. I'll ask you again: «in your opinion is the „write hole” about RAID's „parity” inconsistency only or not limited to?»
Apr
28
comment Software vs hardware RAID performance and cache usage
ok, let's clarify this: in your opinion is the „write hole” about RAID's „parity” inconsistency only or not limited to?
Apr
28
comment Software vs hardware RAID performance and cache usage
As I told you already, there's possibilty that Neil Brown uses term "write hole" more broad than authors of two quotes I cited.
Apr
27
comment Software vs hardware RAID performance and cache usage
Did you ever consider a possibility that developers of LSR meant something more than it's usualy taken when ppl talk about RAID's "write hole"? I gave you two quotes from different sources already. Did you try understanding what they were talking about? Do you understand that LSR's parity "post"-recalc solves the issue those 2 sources were defining as "write hole"?
Apr
27
comment Software vs hardware RAID performance and cache usage
I never said it does it in the same way. Did you read what I said?
Apr
27
comment Software vs hardware RAID performance and cache usage
You were talking about FS metadata now you're talking about RAID's metadata. C'mn, come up with something consistent. I told you that LSR solves "write hole" which is by definition parity inconsistency using parity recalculation either optimized w/ write intent bitmap or w/o it. This is my statement, prove me it's wrong, ah?
Apr
26
comment Software vs hardware RAID performance and cache usage
Ah, nonsence, ah. One question for a braniack: how does cache distinguish between data and meta-data? Good luck.
Apr
26
comment Software vs hardware RAID performance and cache usage
Even for years — it doesn't matter. The only thing which matters whether the cache has complete transaction inside or not. As I told you — think.
Apr
26
comment Software vs hardware RAID performance and cache usage
No, you got me totally wrong. What battery does is only increase the time window duing which data can reach "permanent storage". It doesn't guarantee any integrity by itself. Not a silver bullet at all.
Apr
25
comment Software vs hardware RAID performance and cache usage
blogs.oracle.com/bonwick/entry/raid_z : „… RAID-5 (and other data/parity schemes such as RAID-4, RAID-6, even-odd, and Row Diagonal Parity) never quite delivered on the RAID promise -- and can't -- due to a fatal flaw known as the RAID-5 write hole. Whenever you update the data in a RAID stripe you must also update the parity, so that all disks XOR to zero -- it's that equation that allows you to reconstruct data when a disk fails. The problem is that there's no way to update two or more disks atomically, so RAID stripes can become damaged during a crash or power outage. …”
Apr
25
comment Software vs hardware RAID performance and cache usage
raid5faq.com/reliability.aspx : „… What is a RAID5 write hole? RAID5 write hole is a condition of a RAID5 array when some of the parity blocks do not match the corresponding data. Most often a write hole occurs when a power failure happens during the write operation. In this case a controller has time to write data only but not parity. To avoid accumulating write holes, one should use an UPS (Uninterruptible Power Supply) or BBU (Battery Backup Unit). Additionally, periodical array rebuilds correct accumulated inconsistencies. …”
Apr
25
comment Software vs hardware RAID performance and cache usage
Write hole in regards of RAIDs with parity means parity mismatch due to incomplete write operations occuring. That's why it uses either bitmap optimized or whole array parity recalc. after unclean shutdowns.
Apr
25
comment Software vs hardware RAID performance and cache usage
@shodanshok, you're totally wrong. Think
Apr
24
answered Software vs hardware RAID performance and cache usage
Apr
11
answered OpenVZ MySQL InnoDB won't start
Apr
11
comment OpenVZ MySQL InnoDB won't start
First of all, this is default setting, there's no need to tweak it usually.
Apr
1
comment LVM and snapshots
@James, check out the sources and admit you're wrong.
Feb
20
comment Nginx - 2 https servers on same public ip
David, wouldn't you mind reading serverfault.com/questions/tagged/sni Meanwhile I'm voting to close this q-n
Feb
9
comment How to limit Apache sub-process memory usage?
Yeah, so finally we're inevitably coming to conclusion that Apache doesn't have anything to control such important aspect of resources consumption as memory. )