I'm doing a lot of tests by trying to reload the packages using the scripts that worked before and from the same image and I get always conflicts.
I will report more when I will get more advanced investigation with adrian.
Stef
Hi guys.
Just wanted to mention that when you report problems, please includes the precise steps I need to do to reproduce. When you don't, I waste a lot of time that could be spent fixing things in just trying to second guess what you're doing. Some things I've had problems with: 1. What precise image? there are about 4 6693 image up there at the moment... just give a url to the one you really mean. 2. What scripts to run in what order? 3. Where are the scripts? when I start with some of those image, I simply don't have the scripts you want me to run. So should I first update the ScriptLoader? should I update anything else? please just be explicitly about the steps required to reproduce the problem from scratch.
Remember that since at the moment the process is not standard, nobody except you guys knows exactly how to reproduce things...
Now I want to explore the space thing, so please let me know precisely how to simulate what you guys see.
Daniel
stéphane ducasse wrote:
I'm doing a lot of tests by trying to reload the packages using the scripts that worked before and from the same image and I get always conflicts.
I will report more when I will get more advanced investigation with adrian.
Stef
Hi
I've been using http://ftp.squeak.org/3.9/Squeak3.9a-6693.zip. Was not aware that there are more of those in different versions (I see Squeak3.9a-6693md4.zip and Squeak3.9a-6693md5.zip but I don't know what that is).
Stef, which one did you use?
Daniel, what we did is: - take a 6693 image - load 6694upgrade... - load 6695loadScr...
The two scripts were posted by Stef yesterday to this list. That all works fine for me. Even after flushing the MC caches, my image comes down to 14.3MB.
Marcus tried too and reported: -> It takes a long time. maybe two hours on my powebook -> no conflicts, no merge dialog -> But the image is 19MB large, even after emptying the cache. (I did save the image one before cleaning the caches, as it was unusably slow)
So, also there some difference, at least in the image size.
It would be good if somebody else could run this and report if it works correctly.
It would also be nice if somebody could investigate into improving MC's performance. This would be very welcom,e not only in this case but in general, e.g., we drink too much coffee at netstyle.ch because committing/merging always takes minutes... ;-)
Adrian
On Oct 22, 2005, at 1:08 PM, Daniel Vainsencher wrote:
Hi guys.
Just wanted to mention that when you report problems, please includes the precise steps I need to do to reproduce. When you don't, I waste a lot of time that could be spent fixing things in just trying to second guess what you're doing. Some things I've had problems with:
- What precise image? there are about 4 6693 image up there at the
moment... just give a url to the one you really mean. 2. What scripts to run in what order? 3. Where are the scripts? when I start with some of those image, I simply don't have the scripts you want me to run. So should I first update the ScriptLoader? should I update anything else? please just be explicitly about the steps required to reproduce the problem from scratch.
Remember that since at the moment the process is not standard, nobody except you guys knows exactly how to reproduce things...
Now I want to explore the space thing, so please let me know precisely how to simulate what you guys see.
Daniel
stéphane ducasse wrote:
I'm doing a lot of tests by trying to reload the packages using the scripts that worked before and from the same image and I get always conflicts. I will report more when I will get more advanced investigation with adrian. Stef
On 22 oct. 05, at 14:20, Adrian Lienhard wrote:
Hi
I've been using http://ftp.squeak.org/3.9/Squeak3.9a-6693.zip. Was not aware that there are more of those in different versions (I see Squeak3.9a-6693md4.zip and Squeak3.9a-6693md5.zip but I don't know what that is).
Stef, which one did you use?
Exactly this one.
Daniel, what we did is:
- take a 6693 image
- load 6694upgrade...
- load 6695loadScr...
Exact! This is why this is repeatable
The two scripts were posted by Stef yesterday to this list. That all works fine for me. Even after flushing the MC caches, my image comes down to 14.3MB.
Marcus tried too and reported: -> It takes a long time. maybe two hours on my powebook -> no conflicts, no merge dialog -> But the image is 19MB large, even after emptying the cache. (I did save the image one before cleaning the caches, as it was unusably slow)
So, also there some difference, at least in the image size.
It would be good if somebody else could run this and report if it works correctly.
It would also be nice if somebody could investigate into improving MC's performance. This would be very welcom,e not only in this case but in general, e.g., we drink too much coffee at netstyle.ch because committing/merging always takes minutes... ;-)
Adrian
On Oct 22, 2005, at 1:08 PM, Daniel Vainsencher wrote:
Hi guys.
Just wanted to mention that when you report problems, please includes the precise steps I need to do to reproduce. When you don't, I waste a lot of time that could be spent fixing things in just trying to second guess what you're doing. Some things I've had problems with:
- What precise image? there are about 4 6693 image up there at
the moment... just give a url to the one you really mean. 2. What scripts to run in what order? 3. Where are the scripts? when I start with some of those image, I simply don't have the scripts you want me to run. So should I first update the ScriptLoader? should I update anything else? please just be explicitly about the steps required to reproduce the problem from scratch.
Remember that since at the moment the process is not standard, nobody except you guys knows exactly how to reproduce things...
Now I want to explore the space thing, so please let me know precisely how to simulate what you guys see.
Daniel
stéphane ducasse wrote:
I'm doing a lot of tests by trying to reload the packages using the scripts that worked before and from the same image and I get always conflicts. I will report more when I will get more advanced investigation with adrian. Stef
On 22 oct. 05, at 13:08, Daniel Vainsencher wrote:
Hi guys.
Just wanted to mention that when you report problems, please includes the precise steps I need to do to reproduce. When you don't, I waste a lot of time that could be spent fixing things in just trying to second guess what you're doing. Some things I've had problems with:
- What precise image? there are about 4 6693 image up there at the
moment... just give a url to the one you really mean. 2. What scripts to run in what order? 3. Where are the scripts? when I start with some of those image, I simply don't have the scripts you want me to run. So should I first update the ScriptLoader? should I update anything else? please just be explicitly about the steps required to reproduce the problem from scratch.
Remember that since at the moment the process is not standard, nobody except you guys knows exactly how to reproduce things...
Now I want to explore the space thing, so please let me know precisely how to simulate what you guys see.
read adrian report. I thought that it was clear enough and I when I publish a script I always want to have it reproducable from the latest updated image. I will do a control experiment when I'm back from my short holiday trip.
Stef
Daniel
stéphane ducasse wrote:
I'm doing a lot of tests by trying to reload the packages using the scripts that worked before and from the same image and I get always conflicts. I will report more when I will get more advanced investigation with adrian. Stef
packages@lists.squeakfoundation.org
-
Adrian Lienhard -
Daniel Vainsencher -
stéphane ducasse