Does usr corrupt partition mean replacing hard drives?
Yesterday one of my server went down for more than 20 hours. I complained numerous times to my hosting company and most of the time they replied "we have tweaked apache". Yesterday they said "/usr partition on server has a corrupted file system" I am wondering did that happen to secondary hard drive as well?
After 6 hours I got this reply "We're currently in the process of reimaging your server. " and after one hour I got this reply "I have replaced the disks and reinstalled the OS on your server. "
I am not much expert in Linux environment, usually when something happens in windows partition we try to fix it and it usually gets fixed, is usr partition that much unreliable?
When I asked why you guys are not using secondary drive they replied "There were some file system issues on the original RAID so it is not simply a matter of replacing failed disk and rebuilding the array. "
My question is am I being told wrong? This hosting company is reputable and I have good experience with them so I won't say a word until I find that its not their ignorance.
P.s. I bought this server few months ago, so I suppose to get fresh os etc and this server barely gets 2500 users. It's pretty much loaded server 8 GB RAM, 32 cores and RAID 1.
Here's the thing jcodemasters, when a /usr partition gets corrupted and nobody notice it early, and when it tries to image itself on the secondary drive, it will re-image everything over, including the corrupted portion of it. There's nothing to tell it not to do it because there's nothing wrong with writing an image over.
Since disk 1 had the partition corrupted and I'm guessing some file are effected, they had to correct this problem and check the Disk 2 to see if the problem is there and if not, then they will reimage it up again. If its there then they have to correct it before they could put it back in because if they did, it would ruin Disk 2 as well.
So yes, sometimes the problem with Raid 1 is that if a corrupted file system is not detected early, it might be so bad as to corrupt both disks as well because it's imaging each other and there's no way of not letting it happen. That's the worst thing about Raid 1 configuration.
Dedicated Servers - 100TB Servers - 100Mbps Unmetered Servers - Web Hosting - CDN Network
Servers in Central, East/West Coast USA, EUROPE and ASIA
Welltodo Century - www.welltodocentury.com
Based on your initial reply and the response from the company, it simply looks like the RAID was never activated, as if Raid 1 was in place and one of the drives were to become corrupt in any way, your second one would still contain all of your data. It would be a freak of nature for both of your drives to break in the same exact location at the same time. The Usr partition most likely contains your control panel and software data/configuration, so if they have re-imaged your server, most likely your server is acting like a new one (meaning you would need to set up all of your domains, emails, accounts, etc), though I would strongly suggest asking for a migration to a new hardware or replacing your drives completely if this hasn't been already done. Your /home folder might still be intact, so your actual web files might be there as well as the /var partition which usually holds your database, so this means you haven't actually lost any data, yet it will be hell reconfiguring everything back the way it was. Any server backups you could restore (then update the older files/databases with the current data) ?
█ SlickWebHost::Quality Personal and Business Hosting Solutions
█ USA and UK, CloudLinux, LiteSpeed, R1Soft Shared Hosting
█ USA andUK Virtual Private Servers with Best Connectivity to Europe, North America and Asia
█ SlickWebHost::Non-outsourced 24/7 Support
Based on your initial reply and the response from the company, it simply looks like the RAID was never activated, as if Raid 1 was in place and one of the drives were to become corrupt in any way, your second one would still contain all of your data.
RAID1 only offers protection from hard drive failure, not data corruption. If the data on one drive is corrupted, so will the data on the other drive.
ASTUTE HOSTING: Advanced, customized, and scalable solutions with AS54527 Premium Canadian Optimized Network (Level3, PEER1, Shaw, Tinet) MicroServers.io: Enterprise Dedicated Hardware with IPMI at VPS-like Prices using AS63213 Affordable Bandwidth (Cogent, HE, Tinet) Dedicated Hosting, Colo, Bandwidth, and Fiber out of Vancouver, Seattle, LA, Toronto, NYC, and Miami