Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
Search - "rant hardware server"
-
https://git.kernel.org/…/ke…/... sure some of you are working on the patches already, if you are then lets connect cause, I am an ardent researcher for the same as of now.
So here it goes:
As soon as kernel page table isolation(KPTI) bug will be out of embargo, Whatsapp and FB will be flooded with over-night kernel "shikhuritee" experts who will share shitty advices non-stop.
1. The bug under embargo is a side channel attack, which exploits the fact that Intel chips come with speculative execution without proper isolation between user pages and kernel pages. Therefore, with careful scheduling and timing attack will reveal some information from kernel pages, while the code is running in user mode.
In easy terms, if you have a VPS, another person with VPS on same physical server may read memory being used by your VPS, which will result in unwanted data leakage. To make the matter worse, a malicious JS from innocent looking webpage might be (might be, because JS does not provide language constructs for such fine grained control; atleast none that I know as of now) able to read kernel pages, and pawn you real hard, real bad.
2. The bug comes from too much reliance on Tomasulo's algorithm for out-of-order instruction scheduling. It is not yet clear whether the bug can be fixed with a microcode update (and if not, Intel has to fix this in silicon itself). As far as I can dig, there is nothing that hints that this bug is fixable in microcode, which makes the matter much worse. Also according to my understanding a microcode update will be too trivial to fix this kind of a hardware bug.
3. A software-only remedy is possible, and that is being implemented by all major OSs (including our lovely Linux) in kernel space. The patch forces Translation Lookaside Buffer to flush if a context switch happens during a syscall (this is what I understand as of now). The benchmarks are suggesting that slowdown will be somewhere between 5%(best case)-30%(worst case).
4. Regarding point 3, syscalls don't matter much. Only thing that matters is how many times syscalls are called. For example, if you are using read() or write() on 8MB buffers, you won't have too much slowdown; but if you are calling same syscalls once per byte, a heavy performance penalty is guaranteed. All processes are which are I/O heavy are going to suffer (hostings and databases are two common examples).
5. The patch can be disabled in Linux by passing argument to kernel during boot; however it is not advised for pretty much obvious reasons.
6. For gamers: this is not going to affect games (because those are not I/O heavy)
Meltdown: "Meltdown" targeted on desktop chips can read kernel memory from L1D cache, Intel is only affected with this variant. Works on only Intel.
Spectre: Spectre is a hardware vulnerability with implementations of branch prediction that affects modern microprocessors with speculative execution, by allowing malicious processes access to the contents of other programs mapped memory. Works on all chips including Intel/ARM/AMD.
For updates refer the kernel tree: https://git.kernel.org/…/ke…/...
For further details and more chit-chats refer: https://lwn.net/SubscriberLink/...
~Cheers~
(Originally written by Adhokshaj Mishra, edited by me. )23 -
A few days ago Aruba Cloud terminated my VPS's without notice (shortly after my previous rant about email spam). The reason behind it is rather mundane - while slightly tipsy I wanted to send some traffic back to those Chinese smtp-shop assholes.
Around half an hour later I found that e1.nixmagic.com had lost its network link. I logged into the admin panel at Aruba and connected to the recovery console. In the kernel log there was a mention of the main network link being unresponsive. Apparently Aruba Cloud's automated systems had cut it off.
Shortly afterwards I got an email about the suspension, requested that I get back to them within 72 hours.. despite the email being from a noreply address. Big brain right there.
Now one server wasn't yet a reason to consider this a major outage. I did have 3 edge nodes, all of which had equal duties and importance in the network. However an hour later I found that Aruba had also shut down the other 2 instances, despite those doing nothing wrong. Another hour later I found my account limited, unable to login to the admin panel. Oh and did I mention that for anything in that admin panel, you have to login to the customer area first? And that the account ID used to login there is more secure than the password? Yeah their password security is that good. Normally my passwords would be 64 random characters.. not there.
So with all my servers now gone, I immediately considered it an emergency. Aruba's employees had already left the office, and wouldn't get back to me until the next day (on-call be damned I guess?). So I had to immediately pull an all-nighter and deploy new servers elsewhere and move my DNS records to those ASAP. For that I chose Hetzner.
Now at Hetzner I was actually very pleasantly surprised at just how clean the interface was, how it puts the project front and center in everything, and just tells you "this is what this is and what it does", nothing else. Despite being a sysadmin myself, I find the hosting part of it insignificant. The project - the application that is to be hosted - that's what's important. Administration of a datacenter on the other hand is background stuff. Aruba's interface is very cluttered, on Hetzner it's super clean. Night and day difference.
Oh and the specs are better for the same price, the password security is actually decent, and the servers are already up despite me not having paid for anything yet. That's incredible if you ask me.. they actually trust a new customer to pay the bills afterwards. How about you Aruba Cloud? Oh yeah.. too much to ask for right. Even the network isn't something you can trust a long-time customer of yours with.
So everything has been set up again now, and there are some things I would like to stress about hosting providers.
You don't own the hardware. While you do have root access, you don't have hardware access at all. Remember that therefore you can't store anything on it that you can't afford to lose, have stolen, or otherwise compromised. This is something I kept in mind when I made my servers. The edge nodes do nothing but reverse proxying the services from my LXC containers at home. Therefore the edge nodes could go down, while the worker nodes still kept running. All that was necessary was a new set of reverse proxies. On the other hand, if e.g. my Gitea server were to be hosted directly on those VPS's, losing that would've been devastating. All my configs, projects, mirrors and shit are hosted there.
Also remember that your hosting provider can terminate you at any time, for any reason. Server redundancy is not enough. If you can afford multiple redundant servers, get them at different hosting providers. I've looked at Aruba Cloud's Terms of Use and this is indeed something they were legally allowed to do. Any reason, any time, no notice. They covered all their bases. Make sure you do too, and hope that you'll never need it.
Oh, right - this is a rant - Aruba Cloud you are a bunch of assholes. Kindly take a 1Gbps DDoS attack up your ass in exchange for that termination without notice, will you?6 -
- Launch the new version of the system I have been refactoring for 2 years and counting, then ceremoniously burn (literally) the legacy code as well as the cluster fuck of hardware it runs on.
- Decrease my stress + bus factor by bringing another up to speed on my code & the new version (his cluster fuck now).
- Pay attention to & take better care of health, my wrists in patricular.
- Find a mentor and mentor someone else.
- Get out of crisis management mode and find the time to write tuts, experiment and live a little.
- Find & join a local dev meetup, maybe make a local dev friend.
- Book leave and actually take it, preferabbly without having to take my laptop to the beach - actually, preferabbly at least have the choice to take a offline vacation.
- Sort through the drives containing ALL the code I have ever written, migrate the usefull interesting bits to Github.
Phew, that bit of self reflection was intense! I'm adding a cron to my server to sms & email me this rant in a year to remind me what hope looks like.