| 1 |
On 03/08/2014 22:23, J. Roeleveld wrote: |
| 2 |
> On Sunday, August 03, 2014 10:04:50 PM Alan McKinnon wrote: |
| 3 |
>> On 03/08/2014 15:36, J. Roeleveld wrote: |
| 4 |
>>>> Maybe this "protocol" is not the most clever solution, but it is |
| 5 |
>>>> |
| 6 |
>>>>> one which could be implemented without lots of overhead: |
| 7 |
>>>>> Mainly, I was up to a "quick" solution which is working good enough |
| 8 |
>>>>> for me: If the server has no bugs, why should it die? |
| 9 |
>>>>> Moreover, if the server dies for some strange reasons, it is probably |
| 10 |
>>>>> safer to re-queue the jobs again, anyway. |
| 11 |
>>> |
| 12 |
>>> With the kind of schedules I am working with (and I believe Alan will also |
| 13 |
>>> end up with), restarting the whole process from the start can lead to |
| 14 |
>>> issues. Finding out how far the process got before the service crashed |
| 15 |
>>> can become rather complex. |
| 16 |
>> |
| 17 |
>> Yes, very much so. My first concern is the database cleanups - without |
| 18 |
>> scheduler guarantees I'd need transactions in MySQL. |
| 19 |
> |
| 20 |
> Or you migrate to PostgreSQL, but that is OT :) |
| 21 |
|
| 22 |
|
| 23 |
Maybe, but also valid :-) |
| 24 |
|
| 25 |
I took one look at the schemas here and wondered "Why MySQL? This is |
| 26 |
Postgres territory". It's a case of LAMP tunnel vision. |
| 27 |
|
| 28 |
|
| 29 |
|
| 30 |
|
| 31 |
|
| 32 |
-- |
| 33 |
Alan McKinnon |
| 34 |
alan.mckinnon@×××××.com |
Archives