we are having a problem with webadapter page caches being filled up with different entries generated by URLs which contain gclid parameter which is unique for each URL.
The example of such URL is https://www.a1.hr/INTERSHOP/web/WFS/A1-Shop-Site/hr_HR/-/HRK/ViewStandardCatalog-Browse?campaignID=ljeto&gclid=EAIaIQobChMIiu7U2PGU8gIV2IeyCh311wBOEAEYASAAEgKQW_D_BwE&CatalogID=MobilePhones&CategoryName=MobilePhones
How can we tell page cache to ignore gclid parameter and to cache result for URL without glcid, so that cache key is for example: https://www.a1.hr/INTERSHOP/web/WFS/A1-Shop-Site/hr_HR/-/HRK/ViewStandardCatalog-Browse?campaignID=ljeto&CatalogID=MobilePhones&CategoryName=MobilePhones
In the share/system/config/cluster/webadapter.properies file you'll find some example code:
## Defines a list of query attributes which are to be ignored
## in the page cache lookup. Mainly intended to ignore unwanted
## x/y coordinates, which are included with image button clicks.
# pagecache.ignore.0=submit.x
# pagecache.ignore.1=submit.y
Using pagecache.ignore.n should do the trick.
The knowledge base article is here: https://support.intershop.com/kb/index.php/Display/27984R