Death of InfinityFree


#1

I feel it coming I’m hoping Admin brings back to life our best and most favorite hosting. I been here for a long time maybe even since it launched and this has been the longest downtime that we have had. I hope and have faith that Infinity Free lives long and strong!


#2

LOL, we are with ur feelings.


#3

When did this started?


#4

It has been down for a while, a few days at this point


#5

Yes, truly this is the death of infinity free, all websites are down on unaux and it’s brother partner also, (unaux.com).
Both Infinity Free and ProFreeHost, uses same ftp hostname: ftp://ftpupload.net
Both are going under same issues.
Both have same features, both have same limits, both have same problems… I think that now the living of free hosting culture is going down, no operater is left to provide free hosting. No One!

What will happen in future we all don’t know.


#6

I fell sorry for this community for my home. Can some one tell me any free working hosting provider with same feature. :frowning:

Here is my post link is it same with yours? I have an error is InfinityFree DOWN!?!?


#7

ProFreeHost, but it is also down…
100webhost, with lesser features…
But as far as I know, Infinity Free is the bestest Site ever for free hosting… No other free websites


#8

Byet.host is also down. I have a site running on it, the site runs OK, I can access the control panel but no FTP, no on-line filemanager and no Softaculous. They also use ftpupload.net. As mentioned before it seems ftpupload.net is down.


#9

Have a look at your account/dashboard page. It now shows a message (at last!)


#10

On both https://cpanel.epizy.com and http://panel.byethost.com/ the following message is displayed, so now we know what is happening.


#11

yeah, if you dont consider the system outages


#12

It was all started after migration to new system (I presume by system it means server), from mailing to these errors. InfinityFree’s not dead, they just having hard time because of the new environment. Unexpected behavior upon go-live is very common thing to happen, as I’m sure they already’ve been test the whole system before using it here, somewhere, just not here.

I’m no fanboy of InfinityFree, in fact I’ve migrated to a provider residing somewhere in Asia. I’ve done enough testing stuff, thanks to free service provided here. The old system were good, just not fast to reach wider area. Just gets your hopes up, the team won’t ignore such matters.


#13

Thank you, that’s a very good summary of what’s going on.

There is a reason we don’t use the term “server” so much. Since before I ever worked with iFastNet, they have had a system where all websites are hosted on a big cluster where everything is shared across the entire cluster. If one server went down, it was trivial to bring it up somewhere else with minimal downtime.

However, the setup they were using was aging. There were frequent crashes, capacity issues and performance problems. It’s actually astonishing how few issues there were while iFastNet was working on a new storage system replace the old one, using newer and better technology.

But distributed storage is inherently a complicated subject, and so is the technology used for it. While iFastNet had tested the technology, a small test setup behaves very different from a huge cluster hosting millions of websites. So, inevitably, there were going to be issues which iFastNet could not have foreseen. Anything simple becomes very hard if you make it big, and hosting millions of websites is very big.

The current outage was actually caused by an attempt to improve to storage system, which behaved differently from how they expected it to.

I fully trust the engineers of iFastNet and I know they are doing their best to make the new storage system work better than the old system ever did. But this is complicated technology on a huge scale, so it’s going to take a few attempts before they get it right.


#14

And none of the iFastNet geniuses planned for rollback possibility?


#15

I understand that iFastNet is the upstream provider of our hosting, but “iFastNet has been working on improving [things]” has caused a number of problems during the past few weeks, and for my case it has been practically unusable during the times. Disregarding how a “high load on the storage system” can cause complete outage of the service instead of inconsistent/slow connections, is there a way for iFastNet to inform us before any scheduled maintenance/upgrades that could potentially shut down the whole service (or approximately 1/3 of it), so that we can backup our files, be prepared for it, and plan ahead. Thank you!


#16

I agree there should be a time given where we can back-up everything .


#17

I’m just waiting for this situation to end, I need to test a template for a great project … I do not mind waiting, but to come in fine to be able to use the server


#18

this is beyond cruel and inhumane torture


#19

Yes, me too @Admin


#20

No, because a self rebalancing system can’t be “rolled back”. Sure, you can revert the setting which caused the issue, but then the system will have to rebalance itself to match the old settings, so you’re still going to have the load issues.

That’s the irony of self healing systems: 9 out of 10 times they solve the problem for you. The 10th time they make it much worse.

There have been numerous complaints for months about slow websites and 502 Bad Gateway errors. I’m quite sure that these were the kinds of issues iFastNet was trying to solve.

Because severely inconsistent performance are indistinguishable from a complete outage. If, hypothetically speaking, opening a single file takes an hour, then it would take a day for any piece of software to start. On the system level, the problem is slow storage. On the user level, it’s a service which is unavailable.

Is there a way iFastNet could inform us? Absolutely, there are many ways they could inform their customers proactively. But they don’t. I don’t know why they don’t and I wish they would.

But even if a company does inform proactively about system maintenance, they would only do so for maintenance which they know might cause downtime. From what I know about the root cause, it was just a small configuration change which should not have caused any notable impact.

Also, while you should always have your own backups, I’m curious what you would do with them. Even though websites are temporarily unavailable, no data is lost because of this outage.

cruel [kroo-uhl]

  1. willfully or knowingly causing pain or distress to others
  2. enjoying the pain or distress of others

torture [tawr-cher]

  1. the act of inflicting excruciating pain, as punishment or revenge, as a means of getting a confession or information, or for sheer cruelty

SPEAK NOW OR YOU WILL NEVER SEE YOUR WEBSITE EVER AGAIN!!!

But seriously, if you read anything I have written today, you should already know nobody did this willingly, nobody expected this to happen and nobody is having any fun from this (except maybe competing hosting providers).

Throwing around such rude and obviously wrong accusations don’t contribute to anything whatsoever. Everyone here is doing what they can to fix this issue, and I’ve yet to learn of any issue which was resolved faster because people were being insulted.