New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
I don't know, but I do know that I have had a vps, with basically just a LAMP installed and nothing more, after few month I left it in idle, the login shows like half million of attemptings to log in, a brutal force attack on password.
There always some boring people who scans around and try to hack it.
I updated my PHP to 7.0.20 , if it's not because my site doesn't run on 7.1 ,sure as hell I will make it 7.1.
I updated kernel to 4.11.5 as I checked yesterday it was latest.
I updated mariadb to 10.1.22, if it's not something wrong with plesk I can't figure out, I'm sure I will make it to 10.2.6...
...and such.
1GB RAM is only for the installer, which you don't usually run on a VPS anyway. A minimal CentOS install uses less than 100MB out of the box. And 64bit has been the norm for a long, long time, I don't think many people care about dropped 32bit.
It's really a matter of preference, I just like the underpinnings of RHEL/Fedora more.
...which is pretty intense bloat. Just fired up a 32-bit Deb 8 on Digital Ocean and it's 27MB. That's fresh after install without turning anything off.
So CentOS requires 4x the memory of Debian. That doesn't surprise me.
Deb 7? Under 10MB. I think we had a discussion on LET once about tuning Deb 6 down to < 5MB.
BTW, Deb 8 64-bit was 54MB, so even there CentOS is 2x the memory used.
For CentOS, no, but as I say, their market is not low end boxes.
For LET readers? Lots of 32-bit still in use. Why burn RAM on stuff you don't need, like 64-bit? If you're >4GB of RAM, then yes, obviously, you need 64-bit. But lots of things - e.g., the first three offerings from Digital Ocean or Vultr - are <4GB of RAM and as shown above, the savings by running 32-bit are significant. If I was running a 512 or 1024 (as many here do), I wouldn't even consider CentOS...just wasted RAM.
Heck, I still run some 32MB VPN servers...I'm not going to be running CentOS on those.
Nothing wrong with that, just observing its place in the landscape.
@qtwrk Ugh...Those are port scanners I think.Not much related to your OS version...
@raindog308 where did you get those 32MB servers? thats cool.
I didn't say 100MB, I said "less than" because I wasn't sure about the exact number. But I just checked:
So 47MB, which is less than Debian.
No , i mean in case someone happens to hit you, and happens to OS related security problem and happens on you...
Remember back in May the wanna cry ransom virus?
MS released patch in March , yet still there are countless infections.
And window XP, 7 and 8/8.1 were vulnerable and windows 10 was safe (probably due to the forced update which can't be turn off or just born to immune, I don't know).
But I think it pretty much makes my point about why newer is better I suppose
CentOS 6.9 is still receiving security updates from RedHat and these are still being backported to CentOS. RedHat will provide and support REHL 6.9 and above until 2020, the same date Windows 7 ends support.
More info here:https://access.redhat.com/support/policy/updates/errata (no redhat account required to view)
Do
and then
and then reboot
then do
You should see "CentOS 6.9 amd64" be displayed
If you see 6.8 or below you aren't running the latest kernel, if you see 6.9 you are fully protected and have the latest patches
CentOS 6.9 was released in, well here I'll post a screenshot of the release notes
Link to release notes so you can see it for yourself, https://wiki.centos.org/Manuals/ReleaseNotes/CentOS6.9
@ethancedrik C6 is still fine, so.
Hello,
Maybe the real question is not should i change the OS but should i change the vm ?
If you feel VM is hanging inspect it find out what is eating ressources, tune the system and if it's not possible just upgrade ressources.
You can use commands like #top or #netstat (because each connection eats memory) to diagnose your system.
If you think VM behavior has changed recently remember what were the last modifications made to the system.
You are going off by specific cases that you personally had encountered. I'm sure for one of these "Driver" issue you encountered. There is a lot more incompatible issue for CentOS 7.
CentOS 6 is more than enough for more people. Whatever you are trying to claim is idiotic.
Stop this nonsense rambling.
It's 2017, you can stop using netstat now... iproute tools have been available since kernel 2.2 and net-tools has been deprecated since 2004 at least. Plus ss -m will show actual memory usage of each socket.
Also, /proc/net/sockstat is a good example of why this is probably not a concern, one of my OVZ hosts is using a whopping 3.3 MiB RAM for all sockets. The units in the proc file are the page size.
Good enough if you can deal with ancient versions in their package repos. Any modern software you'll need to build your own RPMs or find some trustworthy repo to provide them, and at that point you're not in supported territory (CentOS project bugtracker will laugh you away) so might as well choose any newer distro. Remember, CentOS 6 packages are frozen in the year 2011. They might get small patches from Redhat but they are stuck in the past for the most part.
I am using centos6 (along with RHEL6 at work) on few servers, surprisingly enough because of hadware compatibily.
Some drivers were dropped from RHEL7/Centos7, and for some older hardware it may be an issue and a real reason to use old OS.
It still works, still receives updates, but i had to compile few key things (like samba for a file server) from source because versions available from repos are ancient.
But other than hardware (or may be some very specific software) compatibility related issues, there is absolutely no point, IMO, to run such old OS. Especially in VM. It creates by far more issues than it is trying to solve.
Pretty sure it's 1 :P