I would consider getting ChatGPT to help you with this if you have not done so already. Keep your items backed up and pay attention to what its telling you to do as to give consent or not, but it can solve a lot of issues in a very short amount of time.
Nope, AI is evil on this forum 
2 Likes
Your funny. By your logic servers, numbers, fire, etc. may be evil too which I totally get that and I don’t think your wrong, but caves can get cold sometimes. I chose a degree in electrical engineering and sometimes I wish I chose acoustic guitar or biblical studies instead. What is it they say in Ecclesiastes all is meaningless under the sun or something along those lines. Definitely personally trying to avoid the robot armies myself so keep to your boundaries if it makes sense to you because it makes sense to me also.
Have you met Joe, he hates AI stuff on this forum, this i what the reference is too
AI hallucinations have been posted here so AI is pretty much banned unless you personally verify the output. I was brewing a batch of beer and was looking up some conversion when the imposed AI answer at the top of the page spit out a conversion that assumed 10 ounces to a pound. And this is the most basic of math based fact. (For you sensible metric folk, the real answer is 16.) So, yeah, at this point if you post AI assisted answers it is on you to verify it didn’t hallucinate and pollute the search engines with further means of circular, self reinforcing hallucination. (As English is the default language here, it is allowed for translation purposes. Even encouraged as even a bad translation is better than none in most cases. Post original language AND translation may not be bad in some cases)
Donning fire proof undies he replies: I think both of those pursuits welcome, in fact, benefit from hallucination. 60’s rock wouldn’t exist with out it. 
I fully expect any reasonable moderator to strike the last part here.
Disclaimer: I’m actually quite the fan of religious experience. I just doubt we differ in the definition.
New rule: Arguing about AI is also prohibited.
The original AI rule is: Don’t post LLM-generated (“AI”) answers to the forum. I’ll delete it when I see them, and I may be grouchy about it. If someone wants to know what the machines think, they can ask for themselves…but, I advise you only follow advice from an LLM after careful testing on a development system.
LLMs are incapable of saying “I don’t know” nor can they ever say “you’re asking the wrong question” (XY problem), and they offer batshit crazy advice sometimes, and more than once we’ve had folks ask for help only after following LLM advice has wrecked their system beyond reasonable recovery. If you’re an expert, maybe talking things over with LLMs is productive, like “rubber duck debugging” where the duck talks back but sometimes makes mistakes. But, if you don’t know what bad advice looks like and you don’t know how to check it, following their advice on a production system is extremely reckless.
But, AI is off-topic for this thread, and that’s the last I have to say about that. I’ve told OP what information is needed to solve the problem (what process or task is spinning up that causes the OOM killer to kick in). They need to give us that information before we can offer useful advice beyond “look at what’s happen on the system when the OOM killer kicks in”.
5 Likes
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.