Los Angeles Free-Net Information

A site to access updated information about the best deal on the Web.

02 December 2015

 
The outgoing mail server is back up and running. There still are a couple of items left to complete to remove the previous blocking. That will occur during the next hour. As best as I can tell, everything is working properly. If you encounter any issues, let me know. Tickets are the best approach at this time.

 
The update to the outgoing mail server is starting as soon as I finish this. Please be advised that although it may appear to be working, any mail you send is likely not to be delivered. Wait until the update is complete and announced here before sending anything you actually want delivered.

15 July 2014

 
The LAFN servers are not accessible at this time. We don't know the cause yet and are investigating. The latest word from AT&T is that we will be backup up in about 30 minutes.

30 May 2014

 
One of the disk drives on the incoming mail server has died. This is a bit unusual since that drive is only a couple months old. We will have to see what we can do to recover from this. It may take some time though. Update: The disk that failed contains the user mail files. Fortunately we have a backup of those that was done early this morning, shortly after midnight. There is adequate space on the remaining disk drive to hold the mail files so I am restoring them from that backup. Thats likely to take about an hour. When that completes we will be backup up and running. There will be some mail you have previously downloaded today in your mailbox. Mail from after the dump was done will not be in the mailbox. To get at that we are going to try and resurrect the failed drive. I have no idea if that will succeed. The logged error messages indicate that is not likely to be successful. It may take a couple of days for us to try that exercise. Update 2: The server is now back in operation. We will be attempting to recover the mail files from the failed drive. Whatever we do recover will be added back into your mailbox.

31 March 2014

 
The servers are back up on the new UPS. We will see what difference that makes. DSPAM and TMDA have now been updated. There will be a bulletin released in a few minutes with the details.

 
The servers are being shutdown at this time to replace the UPS. Should be back up fairly soon.

29 March 2014

 
Something happened this morning. For about an hour before noon today I was unable to access most of the internet include the LAFN servers. I couldn't even get to Google or this blog. I am hearing of others encountering the same problem. I am not seeing anything being identified as the cause yet. I am not sure it affected all users.

 
If you are having problems accessing your mail, your mailbox is probably too large. We have always recommended deleting mail from your mailbox when its downloaded. Now that we are on a slower system, it becomes critical to keep your mailbox size down. Otherwise, the server has to parse the entire mailbox to build an index of the messages for your mail program. If the mailbox is too large, it takes longer than your mail program will wait and it then drops the connection. At that point the server generally leaves over a temporary file because it was terminated abruptly and thus can't be restarted properly. I have to manually remove the temporary file before you can access mail. If you request the temporary file be removed but still have auto download running with your mail program, the problem just returns. The only way to access the mailbox is to use the Mailbox Utility function. Do note, it can't function if the temporary file still exists either. If the mailbox is too large, even it will time out.

26 March 2014

 
There are issues with both DSPAM and TMDA that have been caused by updates to both products and loss of compatibility with our mail server. As a result we are testing a new approach to handling both of them. There will be some changes in how they function. At this time the following appear to the changes. Keep in mind, we are still testing this approach. DSPAM - the addresses you use to report spam or ham (not spam) will most likely become: user_id+spam@lafn.org and user_id+notspam@lafn.org. For example: for me to report spam would be bc979+spam@lafn.org. TMDA - the confirmation process no longer works with our incoming mail server. Actually, I am not all that confident that it worked after the previous system update. We will be eliminating the confirmation request email to the originator. You will still have all the remaining whitelist, blacklist, and quarantine features. You will be the only one who can move a message from the quarantine to your mailbox. There is still a lot of testing required to be sure that mail is being handled properly before we put these into production. In addition, the FAQ will need a major update and we will send out a bulletin with all the changes when they occur.

24 March 2014

 
When we switched to new hardware, the network interface got changed and that didn't get updated in the spam blocking. As a result the drive-by spammers were getting through again. That has been corrected. You should see a significant reduction in spam later today.

22 March 2014

 
It has been reported and verified that there is a problem with DSPAM. Sending an email back as either spam or clean is not working. I can't find anything wrong - it just doesn't work.

20 March 2014

 
The problem with TMDA has been found. We are using a newer version because of the OS upgrade. It requires a additional configuration item that the older versions didn't in order to deliver email. TMDA is now working as best as I can tell. I am not seeing any issues with DSPAM at this time.

 
You can ignore this now, see the entry above. Another bug has been found. If you are using TMDA, either turn it off or remove everything from you whitelist. Mail is delivered to pending properly, but is not delivered to your mailbox. We are working to figure out why this is occurring.

 
One of the side effects of the system upgrade was that one of the authentication files became altered such that only a very few users could authenticate with web functions. Interestingly enough, the ones I used for testing worked. Anyway, that has been corrected. You should be able to access User Services functions now. There was an issue with some of the spam blocking files not being up to date. I believe that has been corrected. However, if you changed spam blocking settings in the last week, you should go to User services and make sure they are still correct.

 
I believe the mail functions are now up and running. Please note that since this is a very small server, it is going to be quite slow. User Services is not available yet. It still requires some checkout.

19 March 2014

 
After replacing the disks, NIC, Memory, and power supply, the server is still unable to run without crashing. As a result we are bringing up a replacement system. Unfortunately, at this time it is a bit light on memory so it may tend to run slower than we would like. I am in the process of recovering the mailboxes at the moment. That may take another hour or so. There will need to be some more checks after that completes, but we should be turning the server back on sometime in the very early morning. There will need to be another outage tomorrow or Friday to add more memory and to physically mount the server in place of the old server. That should only take about an hour at most. There may be some missing renewal notices because some of them were not generated at midnight last night. The server was down at that time. There is likely to be a bunch of email queued up waiting to be delivered to us. We won't be able to catch up real fast.

18 March 2014

 
At this time the disk failures are occurring too frequently. We have only been able to keep the server up for about 10 minutes between crashes. As a result, a number of files become corrupt each crash. To avoid losing any more files, we will probably keep the server down until the new drives arrive. For the couple of users who didn't know (haven't figured out), User Services is hosted on the same server as the incoming mail. As a result it is also down.

 
Now we have discovered a disk drive that is having issues. I am sending replacement drives this morning which should arrive on Wed sometime. Depending on availability of personnel on site, we might be able to start the swap Wed afternoon or otherwise Thurs afternoon. The mail system will need to be down for some time to copy the existing mail files to another system and then bring them back after the new drives are installed. I don't anticipate any issues withe the AT&T router as none of the MACs are being changed. There will be further announcements when we know the schedule.

16 March 2014

 
We now have the router reconfigured and working. You can now download mail (at least while the server is up). We don't know yet if the replacement NIC will resolve the original issue. We have gone as long as 6 hours between crashes so if the problem is resolved we won't know till later tomorrow. There have been two reported file corruption issues from all the crashes. One has been corrected. The other one is a lot more complicated and I don't want to try and restore what we have for it till after we know the problem is resolved.

 
We have identified a failed NIC that is certainly cause some of the problem. That has been replaced, but now we are not able to get the AT&T UVerse modem to pass the IP through to the server. We have this problem anytime anything changes on our network. It will take awhile to work it out. We still don't know if the NIC was the cause of the crashes. Only time will tell on that. In the mean time, however, you can access your email through Remote Mail. The local connections between the servers do work. Unfortunately, no new email is being received since none of the mail servers can get through to our server yet.

15 March 2014

 
Well, the problem is more complicated than it first appeared. With the attack terminated, the system went down again. The uptimes are considerably longer now, but not where they should be (years). We are now investigating hardware issues. However, we may need to upgrade the OS before this gets resolved. That will take a few days. I have an upgraded system in test, but there is still some configuration required and it will take 2 days to send the drives to the facility. At this time we are in a watch and wait mode.

 
The incoming mail server is back up. At this time it appears that the problem was being caused by a large scale attack on the server. That has been blocked and the server has been up for over an hour now, so I believe that was the cause.

 
We are experiencing problems with the incoming mail server. We don't know what the issue is at this point. It does not appear to be a power issue, but there is no indication of a hardware failure either.

16 May 2013

 
Today I am starting the replacement of the news server. It will be going down in a couple minutes and may take a day or two to move all the news data over to the new machine.

Archives

04/06/2003 - 04/13/2003   03/12/2006 - 03/19/2006   03/19/2006 - 03/26/2006   06/04/2006 - 06/11/2006   11/19/2006 - 11/26/2006   12/24/2006 - 12/31/2006   01/14/2007 - 01/21/2007   02/11/2007 - 02/18/2007   05/13/2007 - 05/20/2007   07/15/2007 - 07/22/2007   07/22/2007 - 07/29/2007   08/12/2007 - 08/19/2007   09/02/2007 - 09/09/2007   09/30/2007 - 10/07/2007   10/28/2007 - 11/04/2007   12/30/2007 - 01/06/2008   03/23/2008 - 03/30/2008   03/30/2008 - 04/06/2008   04/20/2008 - 04/27/2008   05/11/2008 - 05/18/2008   05/18/2008 - 05/25/2008   06/15/2008 - 06/22/2008   06/22/2008 - 06/29/2008   07/13/2008 - 07/20/2008   07/27/2008 - 08/03/2008   09/28/2008 - 10/05/2008   10/05/2008 - 10/12/2008   10/19/2008 - 10/26/2008   11/23/2008 - 11/30/2008   02/08/2009 - 02/15/2009   02/15/2009 - 02/22/2009   02/22/2009 - 03/01/2009   05/10/2009 - 05/17/2009   06/21/2009 - 06/28/2009   10/11/2009 - 10/18/2009   11/29/2009 - 12/06/2009   12/06/2009 - 12/13/2009   02/14/2010 - 02/21/2010   07/25/2010 - 08/01/2010   08/01/2010 - 08/08/2010   08/15/2010 - 08/22/2010   08/22/2010 - 08/29/2010   08/29/2010 - 09/05/2010   09/05/2010 - 09/12/2010   09/12/2010 - 09/19/2010   09/19/2010 - 09/26/2010   10/10/2010 - 10/17/2010   11/07/2010 - 11/14/2010   11/14/2010 - 11/21/2010   02/06/2011 - 02/13/2011   07/24/2011 - 07/31/2011   10/02/2011 - 10/09/2011   10/09/2011 - 10/16/2011   10/16/2011 - 10/23/2011   11/20/2011 - 11/27/2011   02/12/2012 - 02/19/2012   01/27/2013 - 02/03/2013   05/12/2013 - 05/19/2013   03/09/2014 - 03/16/2014   03/16/2014 - 03/23/2014   03/23/2014 - 03/30/2014   03/30/2014 - 04/06/2014   05/25/2014 - 06/01/2014   07/13/2014 - 07/20/2014   11/29/2015 - 12/06/2015  

This page is powered by Blogger. Isn't yours?

Subscribe to Posts [Atom]