Intel DG35EC +CentOS 5.x Horrible Write Speed By ICH8 SATA Controller
May 5, 2009
we've been testing CentOS 5.3 on Intel DG35EC board (G35+ICH8+82566 Gb NIC), and found that the write speed out of 7200rpm SATA-II drive, connected to on-board ICH8 controller, is consistently under 10MB/sec which is quite horrible!
the same set of hardware can get 100MB+/sec transfer rate with Debian 5.0 and FreeBSD 7.1, just not the CentOS5.3! it doesn't matter whether AHCI mode is selected in BIOS or not, and of course BIOS has been updated with latest version.
I am looking for a cheap sata controller that can do JBOD with a PCI Express interface(preferably low profile, 2U server) and proper driver support for linux and solaris. The cheapeast one I found is from supermicro but I don't think they can be used in non-supermicro servers. Any suggestions?
I just got 2 dedicateds, and while creating software raid 1, upon initial sync up I'm getting around 7 megabytes per second (6700 kb/s) in write speed I assume. This is a quad core, sata2 setup...
I've got a 'virgin' machine from Nocster running CentOS 5, including a SATA drive (shows up as "SCSI").
It looks like it'll be a straight-forward install [url], but I wondered if anyone has had this exact combination before and if there are any problems I should expect? Given that it's a dedicated machine I don't have physical access to, I'm slightly paranoid about screwing up.
I installed 2 SATA raptors drives on my server. I formatted/partitioned one of the drives through WHM. After rebooting both drives disappeared. They are still detected within the BIOS but not in CentOS. It is possible that I installed a wrong driver or made a bad configuration change.
Server Info:
CentOS 4.6 64bit cPanel/WHM Motherboard: [url]OS is running on a 250GB IDE drive 2 SATA WD Raptors (That I am trying to get to work)
I'm trying to install CentOS 4.4 in dual core 2 box without success, can't even finish installation. Read about a bug closed because it doesn't happen in CentOS 5, thus leaving me no option but to use unstable cPanel 11. Does anyone have any clue about this? Seems strange that there's no workaround.
about the hd,there are two options, the first one is four 7200rpm sata to do raid 10, the second one is two 10000rpm sata to do raid 1, about the performance, which one will be better?
an alternative to the Intel DG31GL board for use in the SC512L-260 1U chassis? Or even an alternative to the DG35EC?
I believe the DG31GL is not available any longer, and I'm looking for something a little more economical than the DG35EC for some budget builds that require only 4 GB RAM.
I'm currently in the process of ordering a new server and would like to throw another $50-$70 at the default SATA II 7k 250 GB to increase performance. The server will host a site similar to WHT (php, mysql, and some forum traffic ).
There are three options I can get for the price:
1. Add another SATA II 7k 250 GB and set up RAID 1 2. Add a 73GB 15k RPM SA-SCSI and put mySQL on it. No RAID. 3. Toss out the SATA II 7k and take two SATA 10k 150 GB instead. Put mySQL on one of them. No RAID.
Please keep in mind that the question is budget-related (I know I can get more if I spend an extra $200 but that's not what I want ). Which of the above will make me happiest?
I have a vps at ifasthosting and it was having problems for the beginning now they completely disabled my account and vps without ANY NOTICE! This creates a big problem for me as I have many live sites on that vps and they are the cause of it!! I never got one email telling me anything or why or what was going on... I have sent them 2 emails in the last hour... What do you guys think I should do?
I obviously need to first figure out whats going on but what after that will I be able to get a back up of my files?
We are hosting our software system which do calculation and file manipulation. Now we have Intel Core 2 Duo E6750 and would like to get Intel Core 2 Quad Q6600 to increase our system performance.
Also we have other choice, get Intel Xeon E5405 Harpertown 2.0GHz to replace two above.
I think it is better to have one server box because one box easy to manage than two. Also Harpertown is much faster than these two together.
I have signed up for wht on good word from a friend that this is "the place to be" when it comes to knowledge of all things web development and hosting. But I also come with sincere warning. If you have come across fastnext.com (which as i have just learned through a mistake in their ticket system used to be hostony) then steer clear. I have been waiting for a response from them on the most ridiculous thing for just over 5 hours now. Apparently when I am trying to download some backup files via ftp the firewall is banning me. They have told me in my ticket that I have now been whitelisted however this does not ring true when I can still not connect to even my mail server. I have had this issue with numerous members who have all been dilligently sending me traceroute reports showing the same thing. My site is a social networking page and apparently when you check your profile too many times or gasp! try to upload or download via ftp. The firewall blocks you. Also they have a ridiculous report to a supervisor link in your ticket that states it will "punish" the tech who is not helping you? I know you guys are from ukraine but what are you going to do to ivan for not helping, send him to a gulag?
I just want you to get me access to my bloody server. Seriously Ive been having issues with them intermittently since day one but not as bad as this morning. And oddly enough I powered down all network equipment on my local area network and forced an ip change through my isp and still cannot get past the 10th hop into chi.fastbighost.com which is not a good thing. I will be moving hosts next month but have not told them this as other stories lead me to believe they may be unprofessional enough to block my access for the remainder of my term therefore hours of work stay on their server unreachable to me. My last backup was before a two day binge of coding work and I just want to get it onto my hard disk to move to a new host. Seriously just do some research and ignore those paid for affiliate sites that have hundreds of "reviews" stating how great they are. dig deeper you will see.
Everything was fine a until a day or so ago.. The server went down and I asked Otto, the owner why it was down. He says it was hacked and the hard drive was corrupted. All my files were lost. So I figured yeah I didn't back up... my fault.. But he said he secured the server when I got it so part was his fault and he admitted it. So he says he'll make a new one with LXAdmin for free as an upgrade since he screwed up... Well now here today, the new server is totally down and I can't access it at all and he's not responding to my messages. So I lost my work for clients and I lost my own personal sites and the money from the people I was hosting. His service was horrible and I wouldn't recommend him to anyone.
My vps has been down all weekend. The stellar support team of Bob, Clide, and Ann has been of no help. They tell me to wait as we look at your problem, the tell me my problem has been escalated, to the higher level of support (who also seem to be Bob, Clide, and Ann), who have told me everything is fixed now (it isn't) and now immediately close any tickets I start, leaving no response.
Is this the normal experience with virpus? Do I need to change hosts?
I signed up for a VPS Land server over a week ago, and everything worked great for the first week. I then began having issues where the server would crash and I would then be stuck at this error:
"System could not log you on. Make sure your User name and domain are correct, then type your password again. Letters in passwords must be typed using the correct case."
Over the last 4 days or so this error has continued to plague me, and after multiple support tickets their only solution was to "Buy more RDP connections" as I was using 2, one for myself, one for my employee to use. I bought 2 more connections for $5 as they recommended and it did NOT fix the problem. In fact, since about Wednesday I've been completely locked out of the server and my support tickets have basically been replied to with:
"We can logon to your box no problem, please email us back if you continue to have issues."
Well, of course, since I'm locked out of the box, I emailed back asking if there is ANYTHING I could do to get it working, as I am losing $200/day in employee fees while he cannot work.
My replies go unheard, and I continue until late Friday night to be completely locked out of the server and my support tickets go unanswered.
I realize that I'm only paying for a cheap $30 a month VPS, but I need this to be reliable. I'm being treated as if I'm completely unimportant and now and I have lost several day's work and employee fees because of this. I even asked if I could pay more to get better service and my emails go unanswered.
I NEED a reliable VPS and cost at this point is no object. With the money I have lost while using VPS Land's HORRIBLE unreliable service I could have had a dedicated box from a real host that actually cares about whether or not it's customers are getting screwed.
I can't logon to the box and they act as though I'm lying.
Pings with 800 byte packets are going through fast and with 0% loss, but http requests, remote desktop, and basically anything more complex than a ping is incredibly unstable and laggy. It can't be a server problem as it's happening on all the servers on a particular network. It's not DNS issues because direct IP connections still have the same issues. What could cause this?
how to contact hostdime's management? Because I can't get any answer that works for me from hostdime hosting and "no" answer from instantcpanellicense..
I am in a very bad situation now.. I did a special agreement with Hostdime's external cpanel seller brand instantcpanel and they don't reply my tickets since 5 days. I am not sure what to do.. My due date has arrived, I need to cancel some licenses, I need some new ones. Also they have removed my special promotional code without telling me!
I have 3 big problems will leave me in some legit sanctions but I am sure they don't care of me. I am willing to pay the problematic payments again Hostdime that's no problem.
I can't believe the hours of BS I had to go through with Godaddy hosting to get a subdomain to work. Their online "help" is horrible, incorrect, and incomplete.
For people who may have this problem, here's what I had to go through:
You can go into "Domain Management" and "Add a Subdomain" and have it point to a specific directory. Fine. That's all well and good, but that in itself does nothing and their help implies that's all you have to do and you're set, boom you have a subdomain. Not so fast.
You have to go into your "DNS Manager" and add a CNAME record with the same name as your subdomain and have it point to @. So apparently you have to change the DNS first so web browsers recognize the subdomain and get redirected to your host, then the domain management "subdomain rule" kicks in and it knows to go to a certain directory.
The funny part is the customer cervice person knew none of this. He kept insisting I had to add an A record, obviously just reading from a script. And even if I did have to mess with my DNS settings to enable my subdomain why wasn't this in the "help" on your website? Unbelieveable.
I have currently a server with a "Intel Core 2 duo e6640 2.4Ghz".
Now the company I'm at moving to a new data center and offering me a box for the same price with a Intel P3.4.
As I understand I would move from a two core to a single core CPU. In the first move letter they send me, they where saying I would get a CPU plus, plus 1 GB mem more.
I recently worked on an issue involving a severe performance issue between "write back" and "write through" caching on the RAID/HD.
Long story short, we purchased 12 IBM x3550 M2s came with LSI SAS1068E/SR-BR10i (gimp redheaded stepchild of the MR series. No BBU, no onboard DIMM.) RAID controller and had very bad and inconsistent write throughput with it. Sometimes it writes out 300-400MB/s (dd test, I know... don't flame. I know dd is NOT a good test.), somethings as low as 30MB/s. The server is configured with 2.5" SATA 500GB on HW RAID-1.
From dmesg log it was default to "write through" on sda. I figured out via lsiutil, you can set the drives to "write back". Once we did that, the write performance is more consistent. NOTE, this enables write back on the SATA drives, NOT the controller itself.
I loop out lspci on all our VPS servers. Found out those with LSI (SAS 8344ELP) cards have sda set to "write through w/ FUA". Those are already all RAID-10s, and I have not heard a single complaint from any customer stating poor I/O performance.
I believe the 8344ELP do have a BBU, I can double-check with DC. The DC is on UPS as well. So that rules out the shortcomings of enabling write back caching.
I want to ask those of you using Xen (3.3 & 3.4). Do you guys have better I/O performance with "write back" or "write through" caching? I'm actually looking for real world results. Where you guys actually deployed production VPS server with clients on it.
one of my servers has 2x250gb hard drives in hardware RAID 0 using a 3ware controller, but ive now got the following error message 3 times prior to 3 server crashes, my server supplier beleves it one of the drives failing (i have backups on a non raid drive so rthats fine) but i suspect the controller, whats everyone else think...
Code: Mar 28 22:41:46 server kernel: sd 2:0:0:0: WARNING: Command (0x2a) timed out, resetting card. Mar 28 22:42:39 server kernel: 3w-xxxx: scsi2: Command failed: status = 0xc7, flags = 0x1b, unit #0. Mar 28 22:43:08 server kernel: 3w-xxxx: scsi2: Command failed: status = 0xc7, flags = 0x1b, unit #0. Mar 28 22:43:12 server kernel: 3w-xxxx: scsi2: AEN: WARNING: ATA port timeout: Port #1.
