By Betty Hardin, Big Blue Water LLC – January 12,
Loop to Loop ..
You’ve setup an email system to send periodic messages to your clients and prospects. You’ve also setup an auto responder so that incoming messages are automatically responded to. Although you have an internal mail server, all of your messages are routed through your ISP.
Your ISP has their system configured to block mass mailings – and blocks those outgoing messages.
Every time a message goes out it gets blocked on the other end and the ISP returns an “unable to deliver” message. The “unable to deliver” message comes into your system and is responded to by the auto-responder. Your system gets stuck in a loop of sending, rejecting, and acknowledging. It loops and it loops and it loops … the system gets slower and slower and before you know it – the server has crashed with 3 million messages in your inbox.
Time required to fix this problem:
• 1 hour to get the server back up and running
• 1 hour to create a temporary mail store
• 1 ½ weeks to recover the ‘good’ mail from the mailbox
• 1 hour to identify the source of the issue (after the mailbox is recovered)
• 3 hours to merge the two mailboxes back into one
When things Blow up ..
You’ve got multiple physical locations with several satellite offices and / or users working from home and on the road. The majority of these people are logging into your system through remote access. You come into work in the morning and the system is slow … the phone starts ringing – you have irate users all over the country. You fiddle around and try several things to resolve the issue. Well into the afternoon you reach your frustration level, then blow up when you call your IT Service Provider – causing a potential relationship issue. Investigation reveals that your Comcast line is running at 2MB – when you are paying for 12MB.
Time to recover:
• 6 hours perusing log files
• 2 hours on the phone with Comcast
• 2 weeks waiting for Comcast to correct the problem
4:00 in the afternoon on Monday..
Things are running slow, something is seriously wrong. Knowing that the last time you had this issue, it was a cable problem you call your ISP to test the lines. They test the lines and find no issues; they’ll send a technician out tomorrow to look closer at things. You do nothing and wait for Comcast.
And then ..
It’s 4:00 in the afternoon on Monday – before you’ve saved a backup for the day – the system goes down. One of the guys in the factory goes into the room or closet where the server is located and turns it off. He tries to restart it. Ooops! it won’t start.
You finally call the IT guys – and they, too, cannot start the system. ALL of the data on your hard disks is gone as well as the mirror image.
Even if you are a one man show …
You are dependent on your computers to be able to do business. Unplanned downtime is not an option.
***ALL OF THESE SITUATIONS COULD HAVE BEEN HEADED OFF AT THE PASS ***
Of course you can pay a person to monitor your systems – at the least you’re looking at an hour of labor once per week. If you want things monitored more frequently than that – it quickly adds up. Manual monitoring requires that the technician has a checklist of things to be monitored and that he diligently checks each of the things on that list. If he doesn’t see something right away or it’s not on the list you can still encounter situations like the ones mentioned above.
What if that guy gets sick, goes on vacation, or is otherwise occupied, what happens?
When a machine is shutdown abnormally it aborts all processes and logging. It may result in damaged hardware and/or corrupt system files. It may not come back up – which would effectively put you out of business until the system is recovered. The logs that normally would have recorded the events leading up to the ‘issue’ most likely recorded nothing. Without logs or screen shots there is no way to know what was truly happening on that machine prior to the shutdown .. which in turn leads to excessive recovery time that neither you or I want to pay for.
While it is necessary to reboot a windows machine from time to time (I prefer this to be a routine procedure), an abnormal shutdown should be avoided– particularly on a server – unless there is no alternative.
When you have a system issue, you will want to nip it in the bud– the longer it goes undetected, the worse it gets and the longer it takes to recover.
We believe the solution is a real-time 24 x 7 monitoring tool.
Call for information at (269) 857-5517