InterWorx 5.0.12 build 567 has been released on the release channel! It’s primarily a maintenance release, and it includes a series of fixes and improvements uncovered since the launch of InterWorx 5.
Below is a summary of the changes included in 5.0.12 Build 567:
Enhancements:
Made the timezone InterWorx uses internally configurable at Server > Settings.
Added support for loadbalancing services imap4-ssl, pop3-ssl, and smtp(587).
[B]Bug Fixes:[/B]
Removed extraneous font files from Heliotrope.
Got rid of c:fakepath in file inputs in certain browsers.
Made MTA default domain a non-required input, since it has no value by default.
Changed minimum subpool size to /126 to alleviate confusion with the feature
Fixed overwriting of custom DocumentRoot settings (if commented out DocumentRoot lines existed further down in config files).
Replaced variables in API output for non-template DNS records.
Fixed adding pointer domains when the working domain's DNS Zone has been manually deleted.
Fixed link on memory graph in NodeWorx.
Fixed display of multiple-ip selector when adding SiteWorx account in heliotrope theme.
Fixed a problem where adding a SiteWorx account on a dedicated IP could cause a command-queue failure.
Fixed display of "host" field when editing DNS Template records.
Fixed one-click toggle of firewall port open/close in NodeWorx.
Fixed a whitespace issue causing some Yum Repositories to mistakenly appear enabled.
Fixed an issue with PTR records failing to respond in certain circumstances.
Improved quickhelp for default nameservers, and made sure that IP addresses cannot validate as default nameserver values (NS Records) for DNS.
Removed Domainkeys remnant of "all/some mail signed" option from DKIM support in SiteWorx.
Fixed display of current working directory in file manager.
Fixed an error triggered by adding a reseller without access to any IP addresses.
Fixed display of e-mail boxes if an underlying e-mail box's folder has been deleted manually.
Disabled senderip on cluster nodes to prevent mail failures.
Fixed an issue with two servers behind the same NAT not being able to setup DNS Synchronization.
Added an option to make messages not be DKIM signed when sent from cluster nodes.
Fixed display of the Cluster > Load Balancer in certain themes.
Fixed an issue with logging into SiteWorx when as a secondary user that is locked to a certain domain.
Many thanks, I’ve installed the upgrade but there’s a display issue on IE, sorry I have not rested it on other browsers, but please see the attached picture. This looks worse depending upon which theme is chosen.
Also, there are some links not correct as follows:
if you goto server, software, there is a manual check for upgrade, this link just takes you back to homepage, and if you goto server, graphs, on our system it shows 2 nics, nic 0 and 1, but if you click nic 1 graph, it displays nic 0 graph. (the graph you have fixed works lovely and displays both nic 0 and 1, from homepage). Also, if you goto reseller, bandwidth, and chose a different reseller, it is displayed in its own box, but only the top few lines are clickable, there is a grey bar covering the rest, which cannot be moved. I have already uploaded a screen shot, bit will do some more tomorrow hopefully.
I hope you don’t mind me posting in this post, but please feel free to move.
Many thanks, I’ve installed the upgrade but there’s a display issue on IE, sorry I have not rested it on other browsers, but please see the attached picture. This looks worse depending upon which theme is chosen.
Find this in ~iworx/html/css/ iwbase.css and delete it. This slipped by us and I’m not sure how.
Also, there are some links not correct as follows:
if you goto server, software, there is a manual check for upgrade, this link just takes you back to homepage, and if you goto server, graphs, on our system it shows 2 nics, nic 0 and 1, but if you click nic 1 graph, it displays nic 0 graph. (the graph you have fixed works lovely and displays both nic 0 and 1, from homepage). Also, if you goto reseller, bandwidth, and chose a different reseller, it is displayed in its own box, but only the top few lines are clickable, there is a grey bar covering the rest, which cannot be moved. I have already uploaded a screen shot, bit will do some more tomorrow hopefully.
I hope you don’t mind me posting in this post, but please feel free to move.
Many thanks, and thanks, you guys rock.
Can you hit me with screenshots of this when you get a chance? Thanks for finding this.
Many thanks and I’ll take some screen shots tommorow if alright, but please see this link for reseller bandwidth issue, and I’ll also check for search function I posted re not working (well on my system anyway sorry).
Hope that’s alright and if you need a tester using IE browser, please just let me know.
Your instructions worked lovely, all is well thank you.
I’m sorry if I seem to highjack this post, and please see pictures for the issues found so far, but I’ll do a little more testing hopefully this week and let you know.
The search function, does appear to work on some now correctly, i.e. lbso.co.uk, when searched using all resellers, does only show lbso.co.uk, but when tested using another siteworx domain, it displays a lot of domains (sorry, domains have been removed as they are live with a reseller account, but if you want to log into our system to have a look, please feel free to do so).
Also, on a off topic, I have resolved the nic start up issue, and when server rebooted, both nics come live with no manual intervention, and if you want, I’ll post or email the process, which I think should be easy to implement for multi network with differing gateways from perhaps a browser in interworx (please note this is for Centos 6.4 64 bit)
I hope this helps, and many thanks for all your help, you guys truly do rock
You saved my ass with finding that Theme glitch so early. I’ve given you a bug-hunter forum award LOL. (Give it a while to recount awards.)
I’m going to look in to some of these and see if I can’t file some as bugs.
Thanks again John. You’re one of the most helpful dudes on the forums.
It’s your support, help and excellent forum that’s makes interworx such a pleasure, and as I said before, we bought the full licence, but were leaving it branded to interworx, as its a brand to be trusted, and all our visitors, resellers and clients see it.
I’ll install later tonight or tommorow morning and run through it, and confirm all’s well of that’s alright.
Once again, many thanks for all your help, it’s appreciated.
Not sure if this is related or just a coincidence, but I’m having some major issues with IMAP today. I shut it down for 10 mins and then start it back up and it’s good for a while, but eventually it pushes my server to a load of like 80 (normally around 0.5). I noticed IMAP mentioned above, so I’m thinking there might be some relation.
I lowered the max connections to 50 for both IMAP and IMAP-SSL for now and also adjusted the /etc/courier/imapd.dist to only allow 1 connection per IP. Does changing this file have any affect on the system or are all the IMAP settings pulled out of some InterWorx Database somewhere?
I’m not an expert sorry but our systems are fine, under the same loading as before and not showing anything out of the ordinary but it’s not clustered, so may not show. We plan on testing clustering very soon, just dependant upon time, we have the test servers and licences etc.
I’d be more inclined to think hardware.
Also, there were a centos kernel update but this may not apply to you if your using a different Linux.
Thanks, I’m just going to keep an eye on it for now.
So far after stopping the service for a little while and starting it back up the second time, it hasn’t happen again. I also did find some strange IPs in the IMAP logs trying to login to a couple accounts on my server, which I just blocked at the firewall,
Any word on “/etc/courier/imapd.dist” actually changing IMAP settings or not? Maybe those changes have helped?
I hope you don’t mind but I was thinking about this, and I think IW does not change the IMAP IP max connections, only the IMAP max connections.
I also think you may have come under attack given what you’ve said.
I would also think setting your max IP connections to 1, is not good. The reason been it may have an impact upon your clients who use IMAP, especially if there are numerous users behind 1 IP, but the number to set it to is dependant upon your overall usage or expected usage of the service. I would probably increase this back to its original default setting, but leave the max imap connections set as you have, unless you receive complaints from your clients.
I would think that the solution to your issue was resolved by blacklisting the offending IP’s, and also rate limiting the connections, after stopping the service for a while (I might have left it 20 mins or so to break sessions though). Lastly, if the offending IP’s are from a known region for undertaking attacks, or have a bad reputation, you may want to consider blacklisting blocks.
I hope you don’t mind, and these are just my thoughts sorry.
Yeah, I’ve bumped the MAXPERIP and MAXDAEMONS back up, but left the NodeWorx connection set at 50 for now.
What’s strange is the IPs I blocked were not from the typical places I’ve blocked before. The 2 I blocked were both from SEVEN network, which seems like a legit operation in California. I’ve since unblocked them and everything is still normal. So now I’m not so sure what the issue was, but just glad it’s resolved.
I do know courier imap can show high resources when 1 or more mail boxes contain more then 10,000 emails or more.
Also, we recently had a specific email issue on our enterprise mailers which was triggered by an email which was opened by webmail of some users, which resulted in very high levels of resource, and was restored by doing a full mail scan and deleting the offending email from the system mailboxes. The email may have been a malware which was not correctly detected but stopped from been fully active by our systems but was still partly opened.
I glad your issue appears resolved and your website looks great by the way.
I’m so sorry I meant to say also, have you checked your logs for pop and smtp, inc SSL versions for the same ip connections, if they appear, they may also show which mail accounts they were trying to access and if shown, you may see its a dictionary or standard word attack, brute force even, and if they are shown, I’d probably leave them in blacklist.
Another good way is to check if the ip are listed in any rbl lists, if so, you may want to review your rbl list checks.
Lastly, you may want to consider honeypot, were signed up and catch quite a few each month, which are new and added to nearly all rbls by honeypot.
I’ve given you a bug-hunter forum award LOL. (Give it a while to recount awards.)
