:apa: スプリットショックウイルス † on Nostr: Linux MD has a Kernel bug that reduces rebuild performance past 16 drives in an array ...
Linux MD has a Kernel bug that reduces rebuild performance past 16 drives in an array significantly, this is resolved in Ubuntu 22.04 by implementing a backport fix from mainline
I believe it
Additionally Raid checksums have also been improved by switching from using cron for scheduling to systemd as it will allow staggering of smaller six hour checksums which can be paused and resumed the next day until the entire array is completely checked rather than checking in one go which can disrupt rsyncs and cause load averages to spike.
I also believe it
Published at
2023-07-06 19:17:15Event JSON
{
"id": "ba044cf499c7c86d77b59af75547f6651c26e0a656c5b40f725cd400ffb80992",
"pubkey": "a1ba0ac9b6ec098f726a3c11ec654df4a32cbb84b5377e8788395e9c27d9ecda",
"created_at": 1688671035,
"kind": 1,
"tags": [
[
"emoji",
"facebook_frog",
"https://mstdn.starnix.network/system/custom_emojis/images/000/071/637/original/8906b71fe17b7798.png"
],
[
"emoji",
"lean",
"https://mstdn.starnix.network/system/custom_emojis/images/000/046/251/original/9ffdcf152a3398a1.png"
],
[
"mostr",
"https://mstdn.starnix.network/users/splitshockvirus/statuses/110668744991966306"
]
],
"content": "\n\nLinux MD has a Kernel bug that reduces rebuild performance past 16 drives in an array significantly, this is resolved in Ubuntu 22.04 by implementing a backport fix from mainline\n\nI believe it :facebook_frog: :lean: \n\nAdditionally Raid checksums have also been improved by switching from using cron for scheduling to systemd as it will allow staggering of smaller six hour checksums which can be paused and resumed the next day until the entire array is completely checked rather than checking in one go which can disrupt rsyncs and cause load averages to spike.\n\nI also believe it :facebook_frog: :lean:",
"sig": "280b668678100e3fdcc1b2f3432c94dbe24fbe9fbd37691958cb5e6858daddeb78c408a22755c683789a618a4aa68ce8a760f3f6732e69d0a336c0029fad7e62"
}
