Email scripts broken last week

S

sanjaysk

Perch
I have been running my asp.net email scripts without any problem for years! Suddenly, they have stopped working last week. The same scripts work on another hosting company without any problem. So something has changed with windows hosting.

I have made a new ticket on this. After following it up over many days, finally I have found that for some reason, the mail servers of the domains are timing out for only the scripts. They do not time out when used through a client program.

The support team keeps saying to use some alternate port. But my point is that these scripts have been running for years. It is better to find out what has broken them--recent changes? Please escalate this ticket to a real expert.

[JH #ZBO-42312-483]
 
Please also consider the fact that many things stopped working in the first week of June (see my recent tickets). Whatever happened is somewhere around June 3 because the last time these scripts worked were on June 3.

Here is a list of things that were broken in this period (June 1st week). This shows that some sort of major change or upgrade was done for my Windows hosting. Perhaps, it will give you some clues.

1) Blogs stopped working due to some mysql and php problem, was fixed.

2) One of my access databases that keep a small log was denied access to the scripts and the permission had to be set again. I will need to see what other things were affected due to the permisison problems.

3) The above script based email problem assigned to this ticket happened around the same time. The last time it worked OK was on June 3.

4) I don't know if this is related but your Cerebrus help desk is not working. I can't see history of any of my recent tickets. I had to send tickets by email only.

Thanks,
Sanjay
 
Sanjay,

upgrade to hosting made....yes rather major, it is called migration. totally changing servers and networks. We've sent out now many mails, dozens of posts about it. The servers are being completely reconfigured per user/domain on new servers in a hybrid cloud setup, on a standardized environment.

It's a pretty big deal, and something you should be quite aware of as we've communicated several times about it, and many many times here on the forums.

There are some things that have happened with this that may make the mail issues you've seen.

The usage of port 587 allows for a faster mail processing when you supply authentication and you should use it all the time.

1. Maybe was php.ini changes, seen this several times as many people still had references to c:\php and other paths that have not existed in production php versions for 3+ years and were phased out in migration.

2. Denied access probably incorrect, but not allowed write access, possibly, as this is something that happened due to the rsync process used to sync files between migrating servers. We reset permissions with a special script but in some locations the DB will remain without write permissions for what it feels are security reasons, as it just does what it is programmed to do, not knowing what your desires may be/have been.

3. Very possibly going cross datacenter and hence the request to use port 587. While it should work in most cases without that it will go through additional reverse lookup checks and the like as part of anti spam processing making a timeout more likely.

4. This can happen if your on file contact address and email address are out of sync, and it sounds like this is possible if you are not aware of the migrations that have been ongoing in a big way.
 
Hi Stephen,

I understand. I thought migration was over a month back. I will recheck the login email for the history of tickets.

Thanks,
Sanjay
 
sanjaysk said:
Hi Stephen,

I understand. I thought migration was over a month back. I will recheck the login email for the history of tickets.

Thanks,
Sanjay
Click to expand...
Are you on cluster2?

That cluster is done, cluster1 is not. I thought from your join date (one of earliest customers) you would be on cluster1 for at least some :)
 
sanjaysk said:
I don't know. How do I find out?

Thanks,
Sanjay
Click to expand...

if you login on cp.jodoshared.com or cp.m****here.biz you are on cluster 1

if on cp.gohsp****.com cluster2

(the **'s are for privacy and google indexing, you should be able to understand what it means)
 
I'm on cluster 1. I will need to try this port setting in the script but that will involve changes to many scripts. I will first test with one script and see how it works.

BTW, I did change my contact email address many months back. So the helpdesk has a bug in that it doesn't display the newer tickets made with that email address. The history displays last ticket from March 10. Please look into this.

Thanks,
Sanjay
 
Stephen said:
The usage of port 587 allows for a faster mail processing when you supply authentication and you should use it all the time.
Click to expand...

Using port 587 solved the problem. Thanks for the convincing reply. I was baffled by the fact that the script had run for years.

Regards,
Sanjay
 
You must log in or register to reply here.
Back
Top