Kevin Foley

  • Content Count

  • Joined

  • Last visited

Everything posted by Kevin Foley

  1. Is there any way to use the AWS Unified CloudWatch Agent and Cloudwatch Logs load EC2 metrics into LogicMonitor similar to what is pulled from the LogicMonitor collector? A little background. We currently use two LogicMonitor collectors to gather Windows metrics from AWS EC2 instances and physical and virtual servers in our on-prem datacenter. This has worked well for the 100 Windows servers and network devices we are monitoring. We are moving all of our on-prem Windows applications to AWS, which requires we change the way we use AWS. Take a look at AWS Control Tower for more details Instead of one AWS account and one VPC ( network ) , we are moving every application and environment ( PROD, UAT, DEV, STAGE ) into different AWS account. No shared networking, so VPC peering or using a TGW isn't an option. This will result in about 20 new AWS accounts and up to 100 by the end of the year. Each of these accounts will have 2 to 20 EC2 Windows instances for web/application along with one EC2 or RDS based Microsoft SQL Server instance. I have no desire to pay for EC2 Windows instances for every account ( possibly hundreds ) just to run the LogicMonitor collector. The most important Windows server monitoring features we rely on in LogicMonitor boil down to - server is down ( ping loss ) - low disk space - high CPU utilization - the ability to alert on Windows Event log entries - Windows service X is not running Then tack on monitoring of an EC2 and RDS Microsoft SQL Server performance data Does LogicMonitor have any recommendations or examples on this type of configuration? Thanks in advance, Kevin Foley
  2. The alerts on the side bar are very helpful when not viewing an alert dashboard or widget. But since it is an alert area, any acknowledged alerts should fall off of the side bar, otherwise it just becomes an annoyance. I can add a filter to the alert display such that it only shows alerts that have not been acknowledged, but this filter is not used on the side panel. I discussed this issue with the LogicMonitor support team and they confirmed there currently isn't a way to do what I am requesting.
  3. How can I monitor host time variance from the collector's time? I have 80+ Windows hosts all of which have time sync enabled to or other ntp servers on my network. From time to time either the ntp client fails, somebody makes a manual time change to a host or some application goes nuts and bends time. Just as I want to be able to alert if a host's disk space is under 5% free space, I want to alert if my hosts clock has skewed 5 minutes from the collector's time
  4. Hello Ryan and Annie, Any update on a simple webhook to notify PagerDuty of alerts? Thanks.
  5. Originally from ticket #13141. In general I want Logic Monitor to always monitor my infrastructure. The exception to this is monitoring websites. I have some single server websites that go through a reset process nightly around 1AM. This process can last a few minutes. During this time I don\'t want Logic Monitor to attempt to monitor the site. Key word is monitor vs alert. SDT would prevent alerting, but still monitor. As far as I know you can\'t even schedule a recurring SDT. Regardless of that my issue is monitoring not alerting. We have noticed that sometimes when the web site is accessed during this reset process parts of the IIS Javascript cache get corrupted. Sure the answer is fix IIS. But the idea of disabling monitoring on a recurring schedule probably has merit elsewhere. Regards, Kevin Foley
  6. Any type of pager duty support would be great. Id like to replace my in-house website monitoring that currently alerts to Pager Duty in favor of Logic Monitors website monitoring, while still maintaining the alerts to Pager Duty
  7. When an SDT is active it only disables alerts, the hosts are still monitored. There are times when I want monitoring turned off during a Scheduled Down Time. Case in point. We run some older Windows 2003k servers with IIS6. Every night we do maintenance tasks between midnight and 2AM. One of those tasks is to recycle the web site application pools. Stopping and starting an application pool can take from zero to 20 seconds depending on server activity. After an application pool is restarted, the first time a web page is accessed the javascript files being served to the client are cached on the server. Due to a bug(s) in IIS6 if these pages are accessed from multiple threads during the website startup and caching phase, sometimes the cache files become corrupted. This means than from 1AM until the time our support staff arrives at 5AM, our websites may be experiencing serious problems. This problem used to happen once every other week. It now happens almost daily and it is causing considerable problems with our clients in different time zones that start working at 3AM. On Sept 18, LogicMonitor added a third webservice monitoring location. Since that time was have noticed almost daily outages on at least one of the 18 sites that are being monitored. It was too coincidental that websites that run on different servers that have not been modified in months all started to exhibit this problem after Sept 18. As a result I am currently disabling all website monitoring. Granted this will not fix the IIS6 bugs when we have heavy activity in the early AM, but it will be one less thing to have to consider as a possible cause. I am sure there are other use cases for disabling monitoring during a SDT
  8. Here is a minor cosmetic request. When setting up a widget dashboard I may have 4 or 10 widgets I want to display, each of varying width. I know that I can position these as I wish, but I would like more control. I may want certain widgets displayed at the top, others in the middle, etc. These widgets may be of different width. Example NOC vs Alerts. Example: Display NOC widgets side by side ( as many as I specify ) , then display the alerts underneath in full width. Here is some pseudo code for a table layout , similar logic could be used for div\'s Widget 1Widget 2Widget 3 Widget 4 Widget 5Widget 6Thanks, Kevin Widget 1Widget 2Widget 3 Widget 4 Widget 5Widget 6Widget 7