Posts made by Visvanath Ratnaweera

Moodle in English -> Hardware and performance -> [Moodle Benchmark] Why is the result of Test 2 good and that of Test 9 bad? -> Re: [Moodle Benchmark] Why is the result of Test 2 good and that of Test 9 bad?

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
Intriguing! Here's an comparison apple to apple:

Dedicated server, Intel(R) Xeon(R) CPU E5-2609 v2 @ 2.50GHz. Linux finds 8 threads. So slightly inferior. Production server. Actively used.

- root file system on SSD:
Disk /dev/sda: 893.101 GiB, 959914704896 bytes, 1874833408 sectors
Disk model: RAID1SSD

There are HD RAID1, for example for moodledata/filedir/.

- 32 GB RAM

The first Benchmark result:

Result of the third try:

First try after Purge all caches:

Third try after Purge all caches:

The difference is huge, as summarized in the Score, 58-66 against 120-160!

I wouldn't call it a Moodle problem because of this control experiment. So my question is again, where is the weak spot in this server.

You might not believe, I'm struggling to get definite answers to what is real, what is virtual. So I will stretch my time accordingly. In the meantime appreciate any insights.

Average of ratings: Useful (1)

Moodle in English -> Hardware and performance -> [Moodle Benchmark] Why is the result of Test 2 good and that of Test 9 bad? -> Re: [Moodle Benchmark] Why is the result of Test 2 good and that of Test 9 bad?

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
Hi Dan

No, I wrote the OP less than an hour after installing Moodle, Moodle 4.1.4+ (Build: 20230804). The only additional plug-in at that time was the Moodle Benchmark. In the meantime I've two more, the Speed Test and the Opcache management. That's all.

What download_file_content() does more than the file_get_content()? The server is weak in that diff.

Background: This is not "my" server. A fairly large academic institution wants gauge how good this machine is for their Moodle. Considering its specifications the results are disappointing. Although it is not the latest model, the smallest RPi should not beat this machine in any discipline! I'm looking in to Why.
Average of ratings: -

Moodle in English -> Hardware and performance -> [Moodle Benchmark] Why is the result of Test 2 good and that of Test 9 bad? -> Re: [Moodle Benchmark] Why is the result of Test 2 good and that of Test 9 bad?

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
Hi Brett

I thought I ran the test three times before I took the screen shot. I rechecked. The screen shot below is taken after 24 h of idle time. After that I ran the test about ten times. The score of 9) improved constantly, went from 1.05 sec to the range 0.69-0.73 sec and stabilized.


Then I purged the cache. Here is the result immediately after that. Rerunning the test brought the same results:

After idling for ~20 min the time of test 9 even went up to 1.64 sec!

Average of ratings: -

Moodle in English -> General help -> Beginner Advice Needed: Group-Based Course Visibility -> Re: Beginner Advice Needed: Group-Based Course Visibility

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
Did you put those disjoint topics in to a single course? That would ring alarm bells in me. The Course is the basic building block of Moodle as such supposed to have a certain character. I would create a course per disjoint topic. If you need an example, you can study how Moodle Academy is structured.
Average of ratings: -

Moodle in English -> Installing and upgrading help -> Error - Apache is functioning normally -> Re: Svar: Re: Error - Apache is functioning normally

by Visvanath Ratnaweera -
Picture of Particularly helpful Moodlers Picture of Translators
No! The .htaccess you found is genuine. Sometimes people get .htaccess files in the Moodle code directory inadvertently. And it easily gets forgotten since "hidden" due to the dot at the beginning.

Sorry, your case is very unusual. Just more random thoughts, they may or may not help:
- Create an index.html file in the Moodle code directory and see whether it responds. Also see whether it makes a difference whether you type index.html in the URL or not.

- Create a phpinfo.php file in the Moodle code directory with the single line <?php phpinfo(); ?> and visit it in a browser. It should bring a very long table about the PHP environment. Look for odd settings in it.

- Study the web server configuration carefully. Are there redirects, proxies or other clever network filters in action?
Average of ratings: -