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.
HostHatch suspended server and ignores ticket
This discussion has been closed.
Comments
we are chill, but this kiddo suddenly talking about cpu steal, in his mind oversold is cpu steal, kid mind. And now he want to argue asking for proof, meanwhile he can't even open google and see the search button in the top right corner, god gives eyes and hands with brain for a purpose.
What do you have against kids?
lmao attack me instead of fact
I never attack you at all what are you with kids? are you dr.disrespect or something?
im out i dont want to talk with dr.disrespect personality
my i m dr never offend
You were going well in the first part.... but then, you dropped the bomb again. I am noticing this again and again about the malicious subtle misrepresentation that you put here and there.
Are you trying to say the customer was not apologetic above?
he even acknowledged his mistake but you again came with subtle agenda above.
You have a very malicious way how you handle things. The chronology of OP and your posts reflecting it again and again.
Hence why he brought up my half-joking thread about cores vs. vCores. It's the second time he does it when it's extremely not relevant, even after saying mods should have closed the thread he wants to continuously bring it up.
I honestly think he deep down likes a little bit of conflict.
Sorry to bring this up again, but about the being malicious part: I must again add that you asked us for a job, did not get it, and that might taint every opinion you have about us, whether positive or negative. Not trying to be an asshole, but again just stating facts. I don't think you are the expert on "maliciousness". It's no wonder to me that you are entertaining the "hosthatch is overselling because they are out of stock" and the "hosthatch is the most malicious person I have seen on LET this year" crowd at this point.
Anyway I think my last response was pretty clear, the trolls will continue trolling and I cannot respond to all of them. Wish you all the best in continuing this crusade.
Why would I be mad about that? I'm really not. I've told you this previously, but you seem to want to continue.
Notice how you've now tried to get me going with two new things, while I haven't thrown anything back.
You like conflict, and that's fine. It's pretty funny how you call this my crusade though after the above, but then again, so long as it feels good, just continue. I'm fairly thick skinned.
@hosthatch by the way, always wondered, the angry man behind your company account here on LET, is it Able or Emil?
I welcome you to read about how bias works. You would be disqualified if this was an actual study. A long time ago.
Its Able. Emil no longer visits LET since quite a while ago, since it gives him too much stress to see all the trolling that goes on here, which is completely understandable. I'd rather have him focus on better things, and the eventual plan is to replace myself with someone who handles better PR, because it gets to me a bit too much sometimes too, especially when children with too much time sometimes have the same amount of voice as actual adults on public forums
Wait, since I don't have any HostHatch servers anymore, dosen't that make you disqualified as well?
Yeah, you're disqualified. Please bring in someone else that can continue to argue with me.
Probably a good idea in general.
Ah, the old "too much time" argument. You must have forgotten that you're here arguing. Unless this only takes time out of me but not out of you? I'm not sure that's how time works.
It's totally fine.
Many providers did so, and many more will do.
@hosthatch if you call this as trolling.... then I actually have no words to say.
This deflection towards straight questions shows how the policies behind the stage are made which does seem ugly.
I don't think it will be too far when we'll see a drama involving @hosthatch
Since I'm getting accused of it, might as well start:
Lol. It actually might be true.
m y i m love lmao
@lirrr you want proof? here is it https://lowendtalk.com/discussion/187216/looking-for-a-dedicated-vds-vps-in-singapore-location#latest
read my comments there. They called it dedicated CPU with 50% stealing
That is what I mentioned earlier. They have enough time to respond here, but not enough time for their clients, who are the source of their income
If that is still happening then it is a shame on @hosthatch with “you should never see CPU steal on a HostHatch VM”
For me, I never encounter cpu steal on my Hosthatch servers even my applications use a lot of cpu.
My VPS went down for 7-8 hours around two months after this happened. So I've moved to OVH. So far, I'm really pleased with OVH's service. Although the panel has some flaws, the server and uptime are excellent. But I'm still unhappy with hosthatch because I opened a ticket that day has received no response as of today. I took their service because someone said they were production-ready. I will not accept it, even if it is free, if I know these things gonna happen. Before then, I had been using their service for a year (it was idle VPS rarely i used it for personal tunneling/vpn), and I had not encountered or observed any problems. When I first started using it on a very busy site, I ran into several issues. first month 50% steal, second month 1 hour downtime (no notification), 4th month (8 hour downtime no notification no ticket response). That day when my server become down I was so worried that each and every moment I was thinking and praying. I was so worried I could not think anything but only about them. when they're going to reply. Are they available? Are they working? Are they on vacation? What if they found it after two-three days? When will my busy site become online again? You know what? If they just say, Hey, we are working on it; that would be enough for me, because at least I know they have acknowledged my server is down and working on it. At least they got my down message. When your busy site is down and you have no idea when your provider will response, that is more dangerous than anything else. That day I learned a lesson, and from that day on I take backups of my database once every minute (only incremental). So that if anything goes wrong i can migrate anytime. Right now whenever i see their name i become angry remembering that day and my situation. They are not production ready, i dont know about other location. But for Singapore they are the worst provider.
Indeed, shame on us if its true. Or shame on the few loud trolls here who have consistently gone off subject, and have continued spreading very clear lies and misrepresentation of facts. In fact, (yes, fact, not my personal opinion), one of the guys who have made most of the negative comments on this thread, is someone who applied for a job and got refused, and has made several negative comments in the past on other threads, and have opened a whole troll thread in the past. To his credit, he has also made some positive comments to come off as neutral. Interestingly enough, most of the negative comments from him have come after that certain event.
I'd recommend reading through the noise, and actually questioning someone if they are claiming something, and then making a more informed decision, based on facts, and not opinions of people who might not like the tone of how I speak (I agree, its not polite, but personal feelings shouldn't change actual facts).
Glad to see, and we plan to keep it that way. In case that changes, you should contact us and we will fix it.
IMHO It is incredibly bad idea. At this frequency you should implement replication for a hot copy of your data somewhere else.
Thanks for your valuable suggestion.
I take the backup and upload it to Google Drive via API; I don't keep it locally. I performed two automated weekly backups (full database) and one incremental per minute (just the most recent rows from two tables) and after two or three month i take a full backup (using rsync to different VPS). DB replication will increase CPU load because simultaneous writing and reading may slow down real-time queries, potentially increasing database response time. I prioritize fast database response, even 1 ms increase in latency matter. Furthermore, I read on StackOverflow that MariaDB replication causes data corruption or create problems in long run. Thus, I have decided to avoid it. I'm running an API, and latency is a significant factor. Taking the backup per minute put very low pressure on my system, As Im using google drive speed is very fast and incremental backup takes around 1-3 cpu seconds to complete thus give very low pressure on the server.
DB Questions (AVG):
ø per hour: 1630800
ø per minute: 27180
ø per second: 553
I have planned for creating multi regional db clustering system. Soon i will implement it. if you have any suggestion please let me know
I haven't worked with high load databases, and I don't know how MariaDB behaves in such cases, but AFAIK making a consistent backup is way more heavy operation then replication. For a consistent backup you would need table locks, which negatively impacts throughput for write transactions.
As for replication I would like to highlight that I mean not master-master scenario but a master-slave one, where the second one is basically hot and almost everytime up-to-date copy of your database. Moreover, you can have frequent incremental backups of it each minute without causing additional load to your main database server that actually handles mix of read and write operations.
Backing up the binlog between db backups could be a middleground, I think?
Sorry, I am not that good in DBA nor know if it is actually a good idea for MariaDB given variety of possible database engines underneath (InnoDB and MyISAM differ a lot).
Pathetic response and attempt at argumentum ad hominem. This whole thread is a really good eye opener for who's behind your 'company'.
Yes, my suggestion was for InnoDB. I only use it. No experience. With MyISAM.
He’s definitely just trying to start an argument, he knows full well that I DM’d them a long time ago, and that I since (after the DM) have been a happy client and ordered new servers. It’s very obvious that I’m not mad about anything related to that, why would I buy more servers afterwards? Again, he knows this. I’m just not pointing it out because he’ll just throw something else in my face to see if that sticks. It’s the common HostHatch response, and he’ll probably do it again even in this thread.
I also posted a good review of them quite recently here on LET, LONG after they didn’t respond to my DM asking if there were any openings. He knows this as well.
Often, he calms down and crafts a few good responses in-between, but is always back to his old ways very quickly.
It honestly sucks because their services aren’t bad at all.
As I'm using INNODB there is no issue of table locking. InnoDB use row locking system thus it creates no pressure at all. MyISAM use table locking which is very bad. In fact nobody can use it in a busy DB. Innodb only lock the rows which is currently updating, since my db only backup the rows which is already updated there is even no row locking issue
The process of mariadb replication is not that much reliable or good if I'm not wrong. first it will create a binlog in local system then forward it to the slave/whatever. which is similar to writing a text file with all the queries then send it to remote server and eventually execute it in remote server.
https://hevodata.com/learn/mariadb-replication-easy-steps/#:~:text=Replication in MariaDB uses a,it to the binary log.
so it's doing the same thing which i'm doing. only difference is its doing real time and i'm doing it after a certain period to reduce the real time load and i'm not executing it to remote server which not only save my money but also headache to manage another server. My process is very simple i'm dumping some rows which updated within last minutes and sending them to google drive via API.
Regarding the load its highly optimized as you can see even after so many queries and millions of requests it's cpu usage is impressive
This server process around 50M request per month. Its a Intel(R) Xeon(R) E-2236 CPU @ 3.40GHz server with 64 GB Ram and NVME disk (raid 1). Thanks to cloudflare
