Reputation
2,024
Top tag
Next privilege 2,500 Rep.
Create tag synonyms
Badges
8 9
Newest
 Necromancer
Impact
~90k people reached

Apr
21
comment Recovering Megaraid Storage Array After Power Outage
@mootmoot, from what I have seen, hardware raid controllers want to wipe out the boot track of the array any time you create it to ensure that no existing data is still recognized, even if you recreate the array exactly the same way as before. If that is all it wipes, then you may be able to carefully reconstruct it by hand though.
Apr
21
comment Recovering Megaraid Storage Array After Power Outage
I've seen this happen to a few hardware raids and is why I no longer trust them. When they screw the pooch, you're SOL. Better to have software raid so you can fix it when things go wrong.
Apr
15
comment SSD Server with trim enabled
@Samul, yes, I who have actually read the ATA standard am stupid because because the all knowing Samul says so; I bow before your superior intellect.
Apr
14
comment SSD Server with trim enabled
@Samul, stop throwing a childish tantrum and pay attention and try to get it this time: it isn't zeroing your data because it is not supposed to zero your data.
Apr
13
comment SSD Server with trim enabled
@Samul, would you also claim that a hammer is not working because it won't screw in a screw? You're using the wrong tool for the job dingus.
Apr
13
answered Can I hot-swap a 2-PSU server from 120 to 240 volts?
Apr
13
answered SSD Server with trim enabled
Apr
1
awarded  Necromancer
Feb
17
comment Why does one of my LAN servers frequently require 21 seconds to connect?
The wireshark log you have shows IP6 failing, then IP4 working. You say that isn't what is going on, but your logs say otherwise. Perhaps you should show a log that actually shows the problem you are describing then?
Feb
16
comment Does a single LACP channel over multiple switches increase redundancy?
@dognose, LACP increases throughput by definition. That is to say: the entire reason for LACP to exist is to increase throughput. You are confusing LACP with the linux bonding driver, which routes packets over multiple interfaces via software and where maintaining the ordering guarantees is difficult if not impossible ( though honestly, who cares about ordering? ). A single interface that has multiple physical ports and hardware support for LACP ( and connected to a switch with LACP support ) does not have such problems.
Nov
21
comment Why dir contents get deleted with chattr +a?
No, as he showed it works fine on a directory and prevents removing files from it.
Nov
2
comment Figure out the used space on the snapshot thin volume
Look at the column labeled "data %"
Nov
2
comment The tool thin_delta segfaulting
Whether it is or is not, the program is not supposed to segfault.
Nov
1
comment Figure out the used space on the snapshot thin volume
lvs will tell you how much space each is using.
Nov
1
comment The tool thin_delta segfaulting
You will need to report a bug to CentOS then, not us. Also segregating has nothing to do with segfaulting.
Oct
27
comment Unable to write to disk but disk isn't full
Actually the mapping is quite linear and reliable. Each error code translates to a specific string when handed to perror(), hence, you get the appropriate error message telling you whether the disk is full, or simply mounted read only.
Oct
26
comment Unable to write to disk but disk isn't full
Nope; been that way for 30+ years. A write to a read only fs returns -EROFS; a write to a full fs returns -ENOSPC.
Oct
25
comment Unable to write to disk but disk isn't full
No actually; when the filesystem has been remounted read only you get an error that states the filesystem is read only, not out of space.
Oct
25
answered Free hypervisor for small configuration
Jul
20
comment Q: MDADM mismatch_cnt > 0. Any way to identify which blocks are in disagreement?
check dmesg or /var/log/syslog?