Configuration parameter reference¶
This document describes each parameter used in sdt.conf
file to configure Synda.
- notes
$HOME is the directory in which you want to place your Synda environment.
Section |
Parameter |
Type |
Default value |
Function |
---|---|---|---|---|
[daemon] |
user |
string |
Set daemon user. The daemon must be started by root for this parameter to work. |
|
[daemon] |
group |
string |
Set daemon group. The daemon must be started by root for this parameter to work. |
|
[download] |
max_parallel_download |
int |
8 |
Set the number of parallel download. |
[download] |
hpss |
boolean |
True |
Gives HPSS service some time to move data from tape to disk. |
[download] |
http_fallback |
boolean |
False |
If HTTP transfer fails, URL is automatically replaced with another HTTP URL. |
[download] |
url_max_buffer_size |
int |
3500 |
limit buffer size as apache server doesnt support more than 4000 chars for HTTP GET buffer. |
[download] |
direct_http_timeout |
int |
30 |
Direct download HTTP timeout, in seconds (time to wait for HTTP response). |
[download] |
async_http_timeout |
int |
120 |
Asynchronous download HTTP timeout, in seconds (time to wait for HTTP response). |
[download] |
direct_db_timeout |
int |
120 |
This timeout parameter specifies how long, in seconds, the connection should wait for the lock to go away until raising an exception. |
[download] |
async_db_timeout |
int |
12000 |
This timeout parameter specifies how long, in seconds, the connection should wait for the lock to go away until raising an exception. |
[download] |
big_file_size |
int |
795795708 |
Files that are a size greater than big_file_size that considered to be big files. These files cannot be downloaded at once => the download is performed into parts (see download.big_file_chunksize parameter). |
[download] |
big_file_chunksize |
int |
16384 |
Big files must be downloaded into several parts to not exceed max memory dedicated to the download process ‘big_file_chunksize’ characterizes the size of each part (i.e. the data size of each http response). |
[module] |
download |
boolean |
True |
If true, download files from ESGF. To use synda in discovery or post-processing mode only, set this parameter to false. |
[log] |
verbosity_level |
string |
info |
Log verbosity level. |
[log] |
scheduler_profiling |
boolean |
False |
Log code profiling information. |
[core] |
metadata_server_type |
string |
esgf_search_api |
Type of metadata server.
Single possible value : |
[core] |
default_path |
string |
|
Default directory path for default selection files. |
[core] |
selection_path |
string |
|
Default directory path for selection files. |
[core] |
data_path |
string |
|
Default data directory path. |
[core] |
db_path |
string |
|
Default database directory. |
[core] |
sandbox_path |
string |
|
Default sandbox directory path. |
[interface] |
unicode_term |
boolean |
False |
Use unicode characters for progress bar. |
[interface] |
progress |
boolean |
False |
Show progress bar for time consuming task. |
[interface] |
default_listing_size |
string |
small |
This parameter drives how many results are displayed by default for “search”, “list” and “dump” subcommands. Possible values are: “small”, “medium” and “big”. |
[interface] |
dump_listing_limit_for_small_mode |
int |
50 |
This parameter drives how many results are displayed by default in small mode for “dump” subcommand. |
[interface] |
dump_listing_limit_for_medium_mode |
int |
100 |
This parameter drives how many results are displayed by default in medium mode for “dump” subcommand. |
[interface] |
dump_listing_limit_for_big_mode |
int |
6000 |
This parameter drives how many results are displayed by default in big mode for “dump” subcommand. |
[interface] |
list_listing_limit_for_small_mode |
int |
20 |
This parameter drives how many results are displayed by default in small mode for “list” subcommand. |
[interface] |
list_listing_limit_for_medium_mode |
int |
200 |
This parameter drives how many results are displayed by default in medium mode for “list” subcommand. |
[interface] |
list_listing_limit_for_big_mode |
int |
20000 |
This parameter drives how many results are displayed by default in big mode for “list” subcommand. |
[interface] |
search_listing_limit_for_small_mode |
int |
100 |
This parameter drives how many results are displayed by default in small mode for “search” subcommand. |
[interface] |
search_listing_limit_for_medium_mode |
int |
1000 |
This parameter drives how many results are displayed by default in medium mode for “search” subcommand. |
[interface] |
search_listing_limit_for_big_mode |
int |
6000 |
This parameter drives how many results are displayed by default in big mode for “search” subcommand. |
[interface] |
show_advanced_options |
boolean |
False |
Allow selection of advanced options associated with some subcommands. |
[behaviour] |
onemgf |
boolean |
False |
Improve search performance. |
[behaviour] |
check_parameter |
boolean |
False |
Perform parameter typo detection (name and value). |
[behaviour] |
ignorecase |
boolean |
True |
Automatically fix incorrect case. |
[behaviour] |
nearest |
boolean |
False |
Automatically select the nearest file replica. |
[behaviour] |
nearest_mode |
string |
geolocation |
Set nearest replica algorithm. Possible values are: “geolocation” and “rtt”. |
[behaviour] |
lfae_mode |
string |
abort |
Set which policies to adopt when a download starts and local file already exists (lfae). Possible values are: “keep”, “replace” and “abort”. |
[behaviour] |
incorrect_checksum_action |
string |
remove |
Set which policies to adopt when checksum doesn’t match. Possible values are: “remove” and “keep”. |
[index] |
indexes |
string |
pcmdi.llnl.gov |
Set the indexes list to use for large operation. This parameter is used for load-balancing on several indexes, to speed up large search-API requests. |
[index] |
default_index |
string |
pcmdi.llnl.gov |
Set the index to use in priority. |
[locale] |
country |
string |
Set the country in which synda is installed. Used to compute nearest replica when “geolocation” mode is used. |
|
[api] |
esgf_search_chunksize |
int |
9000 |
Maximum files number returned by one api call. |
[api] |
esgf_search_http_timeout |
int |
300 |
HTTP timeout in seconds (time to wait for HTTP esgf_search api response). |