How is No Engine Matches a Proxy Issue?

Meatplow

There are dozens of threads on this, but no definitive answer.

I am getting "No Engine Matches" on a list that worked yesterday. I am not searching for sites or anything else. I am only feeding SER a list of verified and working sites for a custom engine I wrote.

I have read that this is a proxy issue? So, I just disabled proxies and it still doesnt work. Yet, I can perform the same task manually from my IP with no issues on all of the sites.

I have tried setting my "page must have" variable to "blank" &  "a|b|c|d|e|f|g|h|i" at separate times, so I know it is matching what is on the page.

page must have1=
page must have1=a|b|c|d|e|f|g|h|i

There has got to be something else going on here? I changed my user agents as well, but still nothing.

Why would these sites return "No Engine Matches" when it does match and I do not have proxies on? There has to be another scenario for why these sites are detecting and giving this error?

gooner

I get that too. No idea what the cause is and there doesn't seem to be a satisfactory answer.

Sven

keep in mind that all "page must haveX" variables have to match. And a a|b|c|d... means that one of those must be present as a word (surrounded by no letter). You might have wanted to write "page must have1=*" which would do what you want.

Anyway it could be a proxy problem if the authorization failed e.g. and an error message is displyed rather than the original content from the URL.

theseo1

One possibility is that the verified URL you're using as a target got deleted and shows an error or redirects to a page not matching the requirements, saw this happen on some platforms.

Still I must admit I'm facing exactly the same problem. Running a huge 200k scrape generates about 5k verifieds in the first run, second run only around 2.200. 50+ semi dedi proxies, fresh email (hotmails same as in the first run) it's just not logical.

I'm running some test over night with the current version of SER and some previous versions to see if that makes a difference (taking the same - unfiltered - list and comparing the results of the current / previous versions). I'll post an update how it went.