Closed timfong888 closed 10 years ago
To do a deploy, I used
java -jar "s3_website.jar" --site public
But not sure if this is missing some of the configurations from s3_website written in ruby....it seems like when running the s3_website command, it wants to load a fresh .jar file into a /var directory....is there a way we could keep the jar file in a specific directory, such as where the gem resides, and it reads from there so I can do a straight download from the source and have it run rather than it fetching with each command?
Here's how you can use te Ruby only version: https://github.com/laurilehmijoki/s3_website/tree/1.x
@timfong888 I'm sorry to hear that you are having a bad experience with this tool. If you let the jar download complete, s3_website should work like before. The jar is about 40 MB.
If you see the "jar file is corrupted" message, it means that the jar download has failed. Slow network connections may cause such failures.
s3_website stores the jar file into the gem directory, if it has a write permission there. Otherwise it stores the jar into the /tmp directory. Do you find this acceptable? If no, how would you improve the download logic?
The jar contains code that is written in Scala. Scala (and JVM) improve the performance of s3_website, because they can access the native threads of the operating system in a way that common Ruby installations cannot. This results in faster push experience for users with large websites. For users with small websites the difference is not significant. The Scala version of s3_website also contains better reporting and error handling.
Ah, that explains it --- I need sudo to give write permissions....I just
checked, that when I run with sudo, it does write into the gem file.
That should solve it! Thanks.
Lauri Lehmijoki mailto:notifications@github.com July 25, 2014 at 8:43 PM
@timfong888 https://github.com/timfong888 I'm sorry to hear that you are having a bad experience with this tool. If you let the jar download complete, s3_website should work like before. The jar is about 40 MB.
If you see the "jar file is corrupted" message, it means that the jar download has failed. Slow network connections may cause such failures.
s3_website stores the jar file into the gem directory, if it has a write permission there. Otherwise it stores the jar into the /tmp directory. Do you find this acceptable? If no, how would you improve the download logic?
The jar contains code that is written in Scala. Scala (and JVM) improve the performance of s3_website, because they can access the native threads of the operating system in a way that common Ruby installations cannot. This results in faster push experience for users with large websites. For users with small websites the difference is not significant. The Scala version of s3_website also contains better reporting and error handling.
— Reply to this email directly or view it on GitHub https://github.com/laurilehmijoki/s3_website/issues/107#issuecomment-50143615.
Tim Fong timfong888@gmail.com Schedule a quick call https://timfong888.appointlet.com
No problem. Please let me know if you discover ways to improve the jar download process or the error messages.
Hi, was wondering when did we start requiring java and a jar file? I haven't been able to upload since I am on the road with low bandwidth. Before, s3_website push was great...now, it keeps trying to load a jar file, even after a successful (and belabored) download. I tried to keep a local copy and run on my Mac so I wouldn't need to keep downloading with every push but it says corrupted....thoughts?
Options for not requiring it or to go back to the latest version that doesn't need it? Thanks much!