We are running loggernet in a PC connected to internet via ethernet. We programmed scheduled data collection to collect data from 7 dataloggers. From time to time (every 1 or 2 days) loggernet crashed with a small window showing the message "Stream error" and no more information. Then I have to manually restart loggernet to collect data again. I have no idea of the reason of that error or what is causing it. Any suggestion? I am getting desperate.
First recommendation is to check that you are running the current version of Loggernet and that the server operating system is up to date. On Windows, that also means the .NET library is up to date. Windows update would check for .NET updates.
If all the software is on current versions, contact your local Campbell Scientific office for help digging deeper. They will want to see a copy of the logs from Loggernet. What was happening right before the error is helpful to see.
Thanks JDavis for your suggestions. All the software are up to date. Where can I find the logs?
The easy way to see the logs is in LogTool within Loggernet. There is an option in the menus to Zip All Log Files. That would put them all into a single file for easy sharing with technical support.
From the logs I found this:
"2022-09-15 19:12:01.792","CR6Irriwell1Router","F","check newest exception","SWC","11577","20220915 17:00:00","Csi::Win32::ReadWriteFile::flush failed","112","Espacio en disco insuficiente."
"2022-09-15 19:12:01.995","PakBusPort","S","received message","src: 1","dest: 4094","proto: BMP5","type: 0x89","tran: 3"
"2022-09-15 19:12:01.995","CR6Irriwell1Router","S","BMP5 message received","type: 0x89","table poll - tempavgs"
"2022-09-15 19:12:02.011","CR6Irriwell1Router","F","Csi::Win32::ReadWriteFile::flush failed","112","Espacio en disco insuficiente."
So maybe the error is due to "not enough space in disk". But this looks a false alert since there free space in the disk I am using. Maybe the problem is that the disk is a network disk, but I don't know why it is falling. I tested the disk copying there a 5gb file and the file was copied successfully, so it is sure that it is plenty of free space. Another tip is that if I close loggernet and open it again it start to work again and can save the data to the disk succesfully. Then, why is happening that alert?
It could be that the network drive at that moment is not available.
in such case, loggernet should wait to the availability of the network drive and retry it later. The main problem is that when the problem occurs loggernet shows a modal window with a single text message "Stream error". Then I can not interact with loggernet, connect, setup... And the only solution is to close loggernet and open it again. That is a critical behaviour for a 24/7 automatic solution.
The files that are referenced in the logs deal with a database of cached data that has been collected from the datalogger and are an integral part of LoggerNet's internal state. It would appear that you have redirected LoggerNet's working directory to be on a network drive. This is problematic for various reasons not the least of which is that network drive connections can be flaky and slow at best. I would recommend storing LoggerNet's working directory on the host computer's local drive and enabling scheduled backups in the setup screen that would write the resulting backup image files to a network drive.
You can also configure data files written by data collection so that these files are redirected to a network drive and, when you do so, you will see that the behaviour is exactly what you describe above. To wit, LoggerNet attempts to open or create a file when it first has one or more records to write and, if that attempt fails, or any attempt to write to these files fails, an error will be reported and the collection schedule will switch to a retry state.
Hi jtrauntvein, thanks for your suggestion. As you mentioned, scheduled backups generate a image file, but I need the .dat files in the network drive, so that solution do not fulfill my requirements.
Sorry, but I don't understand how to apply the suggestion of your second paragraph. Configure data files? How? in which app? Do you mean configure Task Master? I use task master to send the data via sftp to another server. Is it posible to configure it to also copy files to a network drive?
regards,
rAfA
The setting to which I referred can be found in the setup screen under the "Data Files" tab of a selected datalogger. This setting applies on a per-table basis and controls both the name and the path of the data file to be written. What I was suggesting is that you set up LoggerNet to use a local drive for its configuration and state but change the table file settings to write data to a network resource.
i see, sorry I misunderstood your previous message. Then, what you suggest is exactly what i am already doing. But from your suggestions I found a workaround. Instead of save the data directly to the network drive, I could save the data to a local drive and then, using Task master and the "After File Close" trigger, copy the files to the network drive.