Piotr GRD

@wolstech

I use XP and I have a support for SNI-based SSL with no problem in such browsers like Firefox, Chrome, Opera (even the old "real" one v12) or with cURL used with PHP. I didn't check the cURL, but newer browsers (not Opera v12) for default are using TLS 1.2.

However there IS a problem with connecting to SOME websites/servers. I'm not sure why exactly, but probably because of the way the 'handshake' is made or because of how the certificates are verified.

@SMARTDODO

What is a difference for YOU using TLS 1.2 or 1.3 for your connections if SOMEONE ELSE is using TLS 1.1 or 1.0 or is not using encryption at all? Why do you want to force on EVERYONE to use only newest encryption methods?

Remember that forcing newer encryption methods means that you are forcing updates of software. Updating software quite often means forcing the system updates. Forcing system updates quite often means forcing the hardware upgrade or change.

What's for to change the hardware if old one is still working? That's not economical AND not ecological. Even if you will properly in 100% recycle the old hardware (which is currently not tha case) then while transporting it, during the whole recycle process, during production of new hardware, packing it and transporting (very often on very large distance) you are using additional  resources and producing additional trash and pollution.

In my opinion the old hardware should be used for as long as it's possible and that's why I don't really understand this eagerness to update and upgrade everything so quickly with disabling support for older softwares, which means cutting access for everyone that can not or don't want to get new hardware.

And no, security is not an argument for me, because with having old system and old software and computer running 24/7 I have no problems with security since... 2006? And in 2006 that was simply my stupidity and not system or software problem. If someone is conscious/aware computer user than can use older software and have no problems, while if someone is unwise then even yesterday's update won't be enough to protect him.

I can copy my above post from March.

Space left on Tommy's /tmp = 684 kB

Space left on Ricky's /tmp = 52 MB

Space left on Johnny's /tmp = 528 MB

@ sohamb03

It looks that it's using ZeroSSL and also requires being registered.

@ Krydos

When trying to paste LE account key that I've already created in the past with ZeroSSL, it warns me that it's incorrectly formatted. Maybe that's just different prefix?... I'll try to play with it later.

After browsing a little bit, personally - on Windows - I found it easy, quick and comfortable to use Win32/Win64 portable client: https://github.com/do-know/Crypt-LE/releases

It doesn't require any installation, no dependencies, you just need to type the right command.

So it's not web based, I wonder how easy (or not) it will be to instruct some total layman.

I just found that ZeroSSL.com requires now registration to create or prolong Let's Encrypt certificates (no any registration was needed 3 months ago). Additionally you can only get 3 certificates with free account. That was good and easy for layman way to get free certificates, I was able to easily explain what and how to do for non-tech people. Now it's not that cool anymore.

Anyone knows a ZeroSSL-like web based alternative? That doesn't require registration and doesn't limit you to just 3 Let's Encrypt certificates just like ZeroSSL before?

This *is* annoying in cPanel.

But there are two other options (only second one is possible on HelioHost) to have all domains files (including main one) in separate folders.

First option to deal with it in cPanel is to have all addon domains pointed to the folders outside of the main public_html. This is *intentionally* not possible on HelioHost.

Second option is to use .htaccess rewrite engine. Create new "maindomainfolder" folder inside of public_html, move all main domain files in there and directly in public_html create .htaccess with something like:

RewriteEngine On
RewriteBase /
RewriteCond %{HTTP_HOST} ^(www\.)?maindomain\.com$
RewriteRule ^(.*)$ /maindomainfolder/$1 [L]

There may be conflict with other rewrite rules in other .htaccess files inside that folder, some modifications of them may be needed to reflect folder changes.

False results removed. My IP is here, saved with this post.

I think that if - like wolstech wrote - some things didn't work as expected, then that could be also the reason of the ban. Shortly before ban my monitor reported issues with FTP, altough I don't have records what exactly FTP server response was, I only know it was slower and different than expected.

There seems to be no space on Tommy's /tmp.

And by the way, Ricky may need some cleanup of /tmp, too.

Hi, vassbo, and thank you.

My page was designed in 2009 for 800x600 resolution and still looks fine for me in 1024x768.

And for all people using bigger screen resolutions I have always one tip: [CTRL] + [PLUS]. Most browsers remember this setting per domain name. [ctrl] + [minus] for shrinking and [ctrl] + [zero] for reset.

And sorry, but I can't say anything about your style, unfortunatelly, as in my browsers - either in Firefox v40 or Opera v30 (both from 2015 so not old at all in my personal opinion) - whole userstyles org website is simply a blank page for me.

Thank you.

There seems to be some problem with NS2 for over 30 hours already, since Thursday, December 5th 2019, ~14:30 UTC.

For future reference:

In cPanel in the section "Files > Disk usage" you have specified what directories exactly use how many disk space. At the bottom there is whole directory tree, by clicking on the greater-than sign you can browse all sub-sub-subdirectories and find out where exactly your disk space is.

And the disk space available for MySQL/PostgreSQL databases is everything that left from total limit after counting the space used by files and folders.

If someone is thinking about keeping websites online for a really really long term, then always should choose the option to register own top level domain name, which is independent of any webhosting, registrar or any other company.

This way if anything will goes wrong with the webhost - we can move on to another one without changing the address, keep all the visitors and position in the web. If the webhost won't be enough one day and dedicated server will be needed, no problem. Even if the domain registrar will bankrupt, we can change the registrar and keep the domain. At least one own domain for ~10$ a year (to have multiple websites under subdomains of it).

At the very early stage of my "webmaster adventure" I was using free subdomains from the webhosts. Very quickly, after ~2 years I've regret that, when moving between the webhosts I've had to inform all users about the address change. I've moved to free subdomains from the services other than webhosts, which can be pointed to any webhost. That was better solution but again, after ~4-5 years even well known services with good opinions started to disappear. The only one I still know and can be the last resort is freedns.afraid.org (not to mention .tk/.ml/.ga/.cf/.gq domains, which I do NOT recommend at all), all the other that had tens of thousands customers and years of history doesn't exist anymore. So finally one day I've decided to pay for the domain name, at least I know that no matter what will be in next 10 or 20 years with any webhost or registrar, my websites will be still under the same addresses (and I can keep the same email addresses) as long as I'll pay for the domain and have the backup copy of files and databases.

Unless someone doesn't look further than few years ahead, or see no problem in changing the address every few years, then free subdomains from the webhosts are fine.

(...) Anyways that hardly matters took ns1 is up and functional.

@sohamb03

For future reference... In such situations it DOES matter.

If one of the nameservers is not working at all (but the other is working properly) - there is no problem.

If both nameservers are online, but one have no records (or wrong records) for the domain name - there is problem: randomly in ~50% of cases we won't have access to the domain name (website, incomming mails etc.).

No, it's not working properly.

The DNS entry for your domain name is still missing on nameserver NS1, while it's correct on NS2. So still, you - and everyone - will randomly in ~50% of cases access your domain name and in ~50% of cases will not.

Currently it looks that DNS entries for "ricky.heliohost.org" are correct on both nemeservers, NS1 too.

However entries for your domain name are still missing on NS1.

Currently it looks that DNS entries for "ricky.heliohost.org" are correct on both nemeservers, NS1 too.

But the other domain name mentioned in other topic by other user is still missing on NS1. It's possible that there is more such domain names.

It's your decision. Depends where you need it.

You can use .htaccess version in root folder of the domain name (public_html) for all PHP scripts or just in .htaccess inside some subfolder for PHP scripts inside that specific subfolder.

Or you can use PHP version, which obviously will work only inside the script itself (and override .htaccess setting). Most easy is to use one file (configuration file) included in all the other files. But if your scripts/pages are completely separate and have no common part (like for example included configuration file) then in every single .php file where you need it.

Most easy is to put it just in the .htaccess, of course, then you don't have to edit PHP scripts at all, if this is problem for you.

It looks that your domain name is another one (after "ricky.heliohost.org") missing on HelioHost nameserver NS1 (it returns "NXDOMAIN" when asked), while still noted correctly on NS2. That's mean that randomly in ~50% of cases you (or anyone) won't access your website, and in another ~50% of cases everything will looks fine.

We have figured it out already few hours before in the topic mentioned by BlueyEmperor:

https://www.helionet.org/index/topic/36708-cant-access-server-and-transfer-files-via-ftp/?do=findComment&comment=162711

I wrote there that probably there is more such missing domains than just "ricky.heliohost.org" and I was right.

Admin intervention is required, DNS entries on NS1 have to be fixed.

Making changes in .htaccess file (no matter if trough file manager in cPanel or trough FTP) have instant effect (unless there is some cache in your browser for example).

I don't remember since which version of PHP exactly (5.x), but while in older versions setting value of an environment variable "TZ" worked fine, in newer versions you have to use date_default_timezone_set, variable "TZ" won't change anything for you.

# For older PHP versions (up to 5.2 or 5.3):
# - in PHP script:
<?php putenv("TZ=UTC"); ?>
# - in .htaccess:
SetEnv TZ UTC
 
# For newer versions of PHP (function exists since 5.1, but required since 5.4 or so):
# - in PHP script:
<?php date_default_timezone_set('UTC'); ?>
# - in .htaccess:
php_value date.timezone 'UTC'

Most likely it's related to the problem with NS1 (nameserver) which we've just figured out in this topic:

https://www.helionet.org/index/topic/36708-cant-access-server-and-transfer-files-via-ftp/?do=findComment&comment=162711

cPanel, after initial connection without domain name, redirects you automatically to ricky.heliohost.org. For me cPanel on Ricky works, I have logged in. Apart of server being sometimes slow, it's accessible, and everyone of us have 50% of chanses to get the correct DNS response from NS2.

For FTP connection you don't need to provide any domain name, there won't be redirection like with HTTP/HTTPS to cPanel, as FTP host provide the IP and you'll be able to manage files.

Does anyone actually read the error message provided? The one from the browser is most meaningful.

ricky.heliohost.org's server ip address could not be found.

DNS_PROBE_FINISHED_NXDOMAIN

So, this is problem with DNS. I bet that if roguitar would try to connect to "64.62.211.134" instead of "ricky.heliohost.org" (which actually should does not matter at all, there should be no difference in normal situation) then the connection would be succesful.

Let's try to diagnose why there is such DNS problem. Start with the source, HelioHost nameservers, let's check them for example with using my tool HelioHost DNS Checker or any DIG.

Answer from NS2 is correct, but NS1 have no informations about ricky.heliohost.org domain name, it returns the code "NXDOMAIN". That's mean that randomly, in ~50% of cases we won't connect to ricky.heliohost.org. Possibly also some other domain names are missing on NS1.

This needs admin intervention: fixing/synchronising DNS entries on NS1.

 

GRD, When is the page going to be updated? See picture below.

And also, Are you going to make room for Sparkie?

If you already started this offtopic here... (better would be to post in "my" topic about my monitoring, pinned in "customer service").

As jenovaon and sohamb03 wrote already: my page is updated, results of monitoring are uploaded and displayed, but with some delay, not in the real time every two minutes like before. This is because of problems I have since June with the webhost I currently use. Only solution is to move to other webhost (not HelioHost, obviously monitoring of HelioHost should be outside of it), but because of my problems with excessive procrastination it may take "some time" for me. Currently I've managed to set the updates every 30-60 minutes, but with no guarantee, there may be even many hours of delay.

For the same reason - my procrastination - there are no much changes in my monitoring. I think I should check the availability of the server with VPSes (just the mother, not every VPS, of course) and obviously I should get back the monitoring of HelioNet forums and accessibility of the cPanel. There were even plans to add some new tools, slightly changing layout to better fit in various screen resolutions etc., but the same - (very) excessive procrastination stops me. I can only say: sorry.

If required, last 4 posts (snowm's, jenovaon's, sohamb03's and this one of mine) can be moved by moderator to "my" topic about monitoring.

About re-creating hosting accounts with the same usernames as before...

I don't know now, but in the past registration system didn't allow to use specific username (for hosting/cPanel account) if there was already a HelioNet forum account with that name. And because the forum account is created automatically for every new hosting account, then for most of the (deleted?) Tommy's accounts there will be problem with re-creating them with the same username as before.

Solution in the past was to ask the staff to either rename (at least temporarily for the time of registering hosting account) or delete (if not in use) the forum account. Other solution may be to modify registration process, so it won't check the HelioNet if corresponding forum account already exist (unless it's already disabled).

And the other thing, note for the future.

When the maitenance was announced, people asked which servers will be affected. Answer was that Johnny and Ricky won't be. However I think you forget, that during that maintenance both nameservers were offline, so even domains hosted on Johnny and Ricky were affected by that.

Just to remember about this in the future if/when there will be similar situations.

Hi

If disk space consumption will be the same as average for last few weeks, then you have something about a day and a half to do some clean up of the partition /var on Johnny before there will be no available disk space and well known errors start to appear.

Currently there is still 120MB.

Daily average for last three weeks is ~70MB.

Daily average for last seven days is ~80MB.

It's working (well... more or less). With very high server load (~200) for now, so it's extremely slow (or practically inaccessible), but it looks that it's working and is very busy with something.