If you try and view the Umbraco / serilog files that are above a certain size you get the message: "Today's log file is too large to be viewed and would cause performance problems." So is there a way I can decide how big is too big? I've a powerful desktop with 12 cores / 32GB RAM so I'm pretty sure it would be OK with a larger size. When you are doing stuff like trace-level logging it can easily fill up and trying to read a JSON log file in Notepad++ isn't fun... So can we increase the threshold? How? Alternatively, are there any offline log viewers (eg. stand alone app)? I remember someone wrote one once, but was that for the older log files?

@Warren Buckley wrote the compact log viewer you can get it on the Microsoft store (https://apps.microsoft.com/detail/9n8rv8lktxrj?hl=en-us&gl=US) . I use it all the time it's awesome

Yeh use the desktop app & thaks for the plug Kevin

I should have guess it was you @Warren Buckley that's awesome 👍 Thanks @Kevin Jump for the info. I might play around with the Serilog config for dev, but the app should keep me going. Cheers!

No worries

Looks promising , @Mike Chambers 👍

Yep sure can replace it. I think I wrote docs sample to read logs from Azure blob at the time. But yep this is the route to go if you wanna replace that logic/chdck

I'd also advocate maybe using a different logging engine, we turn off file based logging and instead add in logging to SEQ via a serilog sync.

json
"Serilog": {
  "MinimumLevel": {    //Verbose|Debug|Information|Warning|Error|Fatal
    "Default": "Debug",
    "Override": {
      "Microsoft": "Warning",
      "SixLabors.ImageSharp": "Warning"
    }
  },
  "WriteTo": [
    {
      "Name": "Async",
      "Args": {
        "configure": [
          {
            "Name": "Seq",
            "Args": {
              "serverUrl": "https://localhost:5341",
              "restrictedToMinimumLevel": "Information"
            }
          }
        ]
      }
    },
    {
      // not a real sync so can't be in async
      "Name": "UmbracoFile",
      "Args": {
        "RestrictedToMinimumLevel": "Fatal"
      }
    }
  ]
},

As we are talking local dev time here you can have a local instance of seq with improved log functionality as that's its single purpose.. https://datalust.co/ with https://github.com/datalust/serilog-sinks-seq Sql query language for log filtering, and reporting etc... One nice one is the log retention.. set to say 30mins and you machine doesn't fill with logs.... 😉 Also comes into it's own for load balanced setups for aggregating into a single log appliance. You could also use application insights sink too, but for locally logging a little overkill..

Yep I would totally go for SEQ if you can in your setups