Hello,

my hadcm crashed with an error message (again). Just this time the task was gone and a restart of no avail, it is still in the projects folder but no longer displayed in the manager. I was at approximately 7%, so if it\'s irreversibly gone down the drain, I can upload the folder if requested (572kbit upload).

Markus

---

Hi Markus,

It looks like your model crashed with a -161 error. These are hard to diagnose, since the error message basically means \'it crashed, and as a result I have nothing to upload\'.

Take a look at the \'solutions to model crashes\' post (which is usually at the top of the forum) for some ideas on how to avoid this.

But the biggest single suggestion I can make is to take backups of the boinc folder at intervals, when the model is shut down (I do this weekly).

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

Hello Mike,

thanks for your reply, I\'m just afraid it cannot be applied to my problem. The crash was the beginning of my problem, the problem at hand is that the task is no longer displayed in the boinc manager although the data are still in the projects folder. So I would see to avoiding future crashes if the problem persisted and if I still had a task my pc could work on at all.

My offer to upload my folder does not refer to an error message of the manager (cannot upload or alike) but it rather refers to me uploading the data that have been created so far personally to the project if requested and as the progress was at about 7%, they would not be lost and thus the cpu time spent on it would not be wasted.

Again, the task will not resume its work as it is no longer shown in the boinc manager but at least the data still seem to be there.

---

Hi Markus,

The model uploads it\'s data at the end of each model year in any case, so the scientists will already have the information, but thanks for offering :-)

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

Hi Mike,

yes, I know it\'s uploaded. Otherwise I wouldn\'t have been granted any credits yet but according to this

(\"At the end of a model, critical features of interest to the scientists are extracted and stored in 5 smaller files totalling 7MB which will be uploaded....\")

only the results are uploaded but for someone to resume the work unit on my behalf

(\"Archivers Some people are willing to store data if there is a chance that having it available will save the work being recrunched.\")

I assume the whole data are needed. I know it says \"recrunch\" but mine isn\'t nearly finished and so only the \"critical features of interest\" might not suffice to continue the wu. That\'s why my offer and because it is not done in general (\" Climateprediction.net don\'t have the staff or the storage capacity to cope with people sending in all the information from large numbers of BOINC runs.\")

Still my problem persists and the upload would be unnecessary if I could continue it myself.

---

If the model isn\'t in the manager\'s Tasks tab, then it\'s gone for good.
The remanents left in the folder under projects are useless for continuing.
Let BOINC download a new set of values and start a new model.

Thx, clear, precise, short and fast answer. The folder\'s entered manitu\'s eternal hunting grounds already.

But I\'m not sure if there\'s any point to downloading a new model of climateprediction.net. I\'m looking at the Seasonal Attribution Project and I think I\'m giving it a chance.

---

The SAP model is good as a break, I\'m running it now. One thing to watch out for is the long time between checkpoints - anywhere from 40 minutes to 90 minutes (so make sure you have the \'leave in memory\' setting turned on).

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ