Slower performance with ext4

[Christopher Chan]

Mark Kirkwood wrote:
> Christopher Chan wrote:
>   
>> Heh, what do you know? I have been burned by XFS after a powerloss and 
>> got over 4000 zero length files in a postfix queue. No filesystem 
>> corruption, just zero data files. You want to tell me that postfix does 
>> not use fsync? You can guess what I did to the XFS filesystem mounted 
>> for the queue directory. I destroyed it and got ext3 instead in full 
>> data journal mode. Which I repeated on all the other mtas that had a XFS 
>> filesystem for their mail queue.
>>
>>   
>>     
>
> Hmm - not gonna get into trading personal insults , as nothing is to be 
> gained that way.
>   

No. This is not an insult. You are doing others a disservice by spouting 
myths. I am now calling your bluff on how different filesystems behave 
in regards to fsync requests and challenge you to get an authoritative 
answer from any of the developers of XFS, JFS, ext(x) that contradicts 
what I has said.

> You were running this on server grade hardware? or - let me guess - a 
> workstation with cheap sata drives? I have run many instances of mysql, 
> postgres and oracle on *server* grade hardware [1] with xfs for probably 
> the last 7 years and never have *any* data corruption issue in spite of 
> many power outages...
>   

Did you miss my remarks about when you are not using hardware raid + bbu 
cache? You do know that such hardware covers for any short comings in 
filesystems with regards to data consistency and that that is the reason 
for the existence of such hardware?

> regards
>
> Mark
>
> [1] meaning a designated server mobo with eec ram and scsi (or sas) hard 
> drives.
>
>   


A server motherboard that uses ECC RAM and SAS/SCSI hard drives and 
software raid will suffer the same results. You have been spouting 
inaccurate information about filesystem behaviour that will affect those 
who do not have the means to purchase your uber hardware that covers for 
any filesystem's shortcomings with respects to data integrity. Others 
make do with less by having a full understanding of the behaviour of the 
operating systems they run whether it is FreeBSD and softupdates or 
Linux and its various filesystems that support journaling. You can get 
the same data integrity on lesser hardware (motherboards supporting 
ECC-RAM are no longer the realm of 'server' grade motherboards) if 
configured properly.