Quantcast
Channel: Forums | InterWorx CP
Viewing all 900 articles
Browse latest View live

Mass Import Initial Connect form

$
0
0
Not a show stopper by any means, but it seem irregular so I thought I would bring it up.

When on the Mass Import Tool form (/nodeworx/import/transfer) I fill in the IP, root user, password, select IW.
Then when hitting submit the pages reloads, but basically nothing happens (the form stays filled out, password and all).
Clicking the submit button a second time gets it to work.

Thanks John (aka d2d4j) for keeping the forums lively!

$
0
0
John,

The forums used to be more active 5+ years ago, but as InterWorx has gotten better and better, plus it being so intuitive, there seems to be less activity in here.
But I know when I ask a question, you are here to answer it, or at the very least point me in the right direction.

Thanks for all you do and look forward to seeing you in the forums in 2017.

Happy New Year to you and the rest of the interworx family!


- Justin

Possible issue with IMAP and IPv6

$
0
0
So I say this is an issue because errors are logged in the imap4 log file, but it doesn't seem to break anything.

Here is a snippet of my log file with just the IP address and emails changed:
Code:

tcpserver: pid 6794 from 2601:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx
tcpserver: ok 6794 server.myserver.com:2607:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:993 :2601::xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx::58652
INFO: Connection, ip=[2601::xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx]
vmysql: error creating table 'relay': Table 'relay' already exists
vmysql: sql error[7]: Data too long for column 'ip_addr' at row 1
INFO: LOGIN, user=bob@aol.com, ip=[2601::xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx], protocol=IMAP

When an IPv4 connects it looks like this:
Code:

tcpserver: pid 8727 from xx.xx.xx.xx
tcpserver: ok 8727 server2.justechnology.com:::ffff:yy.yy.yy.yy:993 :::ffff:xx.xx.xx.xx::59326
INFO: Connection, ip=[xx.xx.xx.xx]
INFO: LOGIN, user=jim@thetacoshak.com, ip=[xx.xx.xx.xx], protocol=IMAP

Main A Record

$
0
0
Why was it decided that SiteWorx users would not be able to edit the website's main A record? As when you do click on the edit button the IP address is not in a box that can be edited. I know that anyone with reseller access/Nodworx access can go into the DNS zones part and edit it from there but the customers in question don't have that access and just SiteWorx.

I assume this can't be changed and was designed to be like this? It's just a bit odd and annoying when we have some clients who have hosting elsewhere but emails and other services with us and want to be able to manage the DNS with ease.

Qmail Sender IP and email forwards

$
0
0
I have a client that still uses AOL and have a lot of their email forwarded from their domain to their AOL account. After moving to a new server they were having more emails not make it to AOL. I discovered an issue (typo) that my hosting company had in the rDNS for my new IP, the main IP of the server.

So this got me thinking, does the Qmail Sender IP only work when a user is sending an email themselves, manually?
But when an email is forwarded by the system, it still uses the main IPv4 (or IPv6)?

IPv6 Firewall

$
0
0
I've looked on the site, but haven't found any guides for this.

With the IPv6 firewall set to "Managed by InterWorx" and "InterWorx-Managed ICMP Settings", is there anyway to white list IPv6 addresses? I know because of the magnitude of the IPv6 ranges, it might not be as simple as the Trusted and Blocked IPs in the control panel for IPv4, but just not sure where to start.

Need better management of IPs for Mass Import (iworx to iworx)

$
0
0
I just moved to a new server last week and I've run into a few issues with some IPv6 sites I had setup. On the import it just pulled from the pool (same pool as on previous) server, but pulled different IPs and not matched up to previous ones. So i've had to manually delete the IPs, releasing them to the "static pool" of IPv6 and then assign them to the right site (or change the DNS).

I have run into this before on IPv4, but with just a few dedicated sites and the rest on a single shared IP it wasn't much of an issue. But static IPv4s also weren't able to match up with the sites by domain (keeping same IPs on both server and just having them routed differently in the datacenter). The static IPs were just handed out in order of how they came up on the import list, for any sites that had SSL setup on the previous server. No big deal with just a few sites, but once I go more to using IPv6 it will be an issue the next time I upgrade servers.

Would be nice if there was an option to check off (or just by default) if the IPs on both servers are the same on both server, where it will use the same IP for the same SiteWorx account (IPv6 and IPv4)

I know this sounds a lot simpler on paper than real world. I'm sure there are a 100 things that I'm not thinking of that would make this more complex, but I have to believe it could be pulled off by the interworx team.

Mass Import Tool: None tgz option?

$
0
0
Lots of new ideas since just recently moving to a faster server and doing an interworx to interworx mass import!

During this move I had one site that was almost 100GB just about split 50/50 between web and email that just wouldn't play nice with the import tool. I tried a few times to do this import (just this one by itself to minimize load vs the two at a time it can do) and it failed every time. So what I wound up doing is a structure only mass import to create the SiteWorx account than then doing a remote rsync copy of all the account's files from the old server to the new one. What took hours, more than 6, to do the full import (which failed), I was able to accomplish in about 1 hour or so doing the remote rsync copy.

Keep in mind this was for moving all my sites from one server to another, inside the same datacenter, plus being able to use a LAN connection between the two servers. I know this isn't the case for everyone, so obviously this would be an option, in addition to the default tgz setup...but would be really cool if it was an option.

My Idea
Basically I'm thinking if you have a situation like mine, a direct rsync between the two servers versus creating a tar package, zipping it, then sending it as one big file to the new server can work much better. You would think with the greater overhead of doing file by file versus just moving one big tgz wouldn't make sense, but in my case it actually worked and was at least 6 times faster.

I understand the tgz method makes sense because it is using the already existing backup system that is used through interworx, but I think this special case would make sense for the mass import tool. Create a simple structure backup, move that over, create the siteworx account on new server (just like it does now), then just rsync over all the files.

ssl.conf CentOS 7 issues

$
0
0
For some reason when I try to update the ssl.conf SSLCipherSuite, to just add on :!RC4 it breaks apache.
It just won't restart correctly. Has anyone else run into this? What could I be doing wrong. On my old CentOS 6 server it worked fine.

CentOS 7 - SSLCertificateChainFile directive is deprecated? Should we change this?

$
0
0
This warning comes up when looking at the apache startup status:
The SSLCertificateChainFile directive (ssl.conf) is deprecated, SSLCertificateFile should be used instead

Should this be SSLCertificateChainFile be swapped out for SSLCertificateFile?

Force /webmail, /roundcube, etc. to SSL

$
0
0
If you try to go to SiteWorx for a domain it will force you to SSL on port 2443. But if you try the same with webmail, it allows you to connect using HTTP. Is there a way to add this same redirect for /siteworx to all the webmail paths?

Also, since most sites won't have SSL can it be redirected to a main domain I have an SSL for?

Upgrading to latest 2.4.x Apache on CentOS 7

$
0
0
Trying to see if I can make my server pass a PCI Compliance test and I'm getting a warning about the Apache version 2.4.10. Looks like it's up to 2.4.25 now. Is there some interworx support yum repository we can use to upgrade Apache?

NodeWorx SSL config doesn't seem to match up for IMAP & POP

$
0
0
On the NodeWorx page /nodeworx/ssl, when editing the SSL Cipher Suite for IMAP and POP, it doesn't seem to have any effect on the underlying config file on the system.
Unless I have the wrong config file, but don't believe I do.

Code:

/etc/courier/imapd-ssl
/etc/courier/pop3d-ssl

Is it possible to force imapd-ssl and pop3d-ssl to TLSv1.2?

$
0
0
From the config file in /etc/courier

#NAME: TLS_PROTOCOL:0
#
# TLS_PROTOCOL sets the protocol version. The possible versions are:
#
# SSL2 - SSLv2
# SSL3 - SSLv3
# TLS1 - TLS1

TLS_PROTOCOL=TLSv1.2+ (This was TLSv1+, can TLSv1.2 work here?)

##NAME: TLS_STARTTLS_PROTOCOL:0
#
# TLS_STARTTLS_PROTOCOL is used instead of TLS_PROTOCOL for the POP3 STARTTLS
# extension, as opposed to POP3 over SSL on port 995.
#

TLS_STARTTLS_PROTOCOL=TLS1.2 (This was TLS1, can TLS1.2 work here?)

IMAP Port 143 with STARTTLS and Outlook 2010

$
0
0
I'm a little behind the game here as I've been using the explicit port 993 for my secure email, but when securing my server for PCI compliance I decided to see how the STARTTLS worked.

I'm running Outlook 2010 on Windows 10 and set the port to 143 and TLS as the security option.
When I try to connect I just get an error:
Quote:

Log onto incoming mail server (IMAP): A secure connection to the server cannot be established.
If I put it to port 993 and TLS:
Quote:

Log onto incoming mail server (IMAP): The operation timed out waiting for a response from the server.
If I put it to port 993 and SSL, everything works as expected.

Is this an issue with the server or Outlook being the old 2010 version?

Has anyone got STARTTLS to work for IMAP?

$
0
0
I thought I had an issue with Outlook 2010 being old, so as a test I got the latest version of Thunderbird which clearly supports STARTTLS.

I changed my server back to TLSv1 temporarily for both TLS and STARTTLS connections and was not able to connect using STARTTLS on port 143.
If I changed it back to the standard TLS on port 993 it works fine.

So trying to figure out if there is a STARTTLS issue on my server, my computer (Windows 10), or something with STARTTLS in general on interworx CentOS 7 servers???

I mean I assume most people, including me, still just use the old way of having explicit ports for secure channels, but this should work as it's enabled on the server by default.

Cloudflare Plug In Issues

$
0
0
Does anyone know if you can reset the Cloudflare plugin? I had my sites using the Cloudflare plugin and needed to change my credentials with Cloudflare's partner site. I deleted the sites and logged out on the Cloudflare plugin, but each time I try to add them again, they report already being registered with Cloudflare. I contacted Cloudflare support and they are reporting the sites are deleted and not active. I need to somehow delete the info being held by the Cloudflare plugin, or uninstall the plugin and re-install it.

Anyone have these issues, and is it possible to uninstall the plugin and re-install it fresh deleting any residual login information being held by the plugin?

Thanks

Optimization work with MySQL

$
0
0
Dear all!

I am stuck in a situation where MySQL greatly slows down the server.


Mysqltuner utility Recommendations only aggravate the situation. The author of the script and I do not change anything in it. I read somewhere that if two users are working with the database, it turns out much better than one? Is it true?

Where is it posible to see DKIM records and where can be changed in INTERWORX

$
0
0
Hi friends ...

Can anyone help us ?

We have problem with our sent mails to gmail ... lot of mails are sent to SPAM.

There is possibly problem with SFP and DKIM records ..


We have corrected SFP record but where is it posible to see DKIM records to be reviewed and where can be changed in INTERWORX

Tanks

JURAJ

Change the log retention for system services like IMAP

$
0
0
I am planning on killing non-secure IMAP4 service on my server. So I ran this command in the log file location:
Code:

grep -i 'INFO: LOGIN' * > imap4user.log
This gave me a list of all the emails that have logged into regular IMAP. Pasting that into Excel I was able to create a nice, no duplicates list pretty quickly.

But my log files don't go back that far, so I was wondering if I could change this so I could get a wider range of emails, then after I could set it back to normal log retention.

This way I can alert my user that they need to switch to SSL.

Thanks!
Viewing all 900 articles
Browse latest View live