hdeadman

I had just imported an updated list, over-writing and adding to an old list, then I sent out a newsletter to the new list. A day or two later I was told by someone that not everyone had gotten the newsletter. I looked in the Email Queue and there were about 300 e-mails still in the queue from a list of about 400 e-mails. There was an error message on on the e-mail queue indicating that a particular e-mail address was invalid. The e-mail address (a gmail address) started with a number 1 so I guessed that might be why. I tried getting the queue to resume by clicking run now but it didn’t seem to be going so I removed two e-mails from the list that started with numbers and it started sending again (maybe I clicked run now again). The e-mail it said was invalid was an e-mail that had been used in previous mailings and was an address on the list before I re-imported the list. It did happen be the first e-mail in the list when sorted alphabetically (because it started with 1). The mail server I am pointing at is an Amazon SES smtp server. I am fairly certain that the error message on the queue page said that the e-mail address was invalid or illegal and the error seemed to stop the queue from being processed.

Does the plugin consider e-mail addresses that start with a number as invalid? I got an error that said an e-mail starting with a digit was invalid but it is a valid address and it worked in the past with this plugin.

Regarding the apostrophe causing a SQL error, that is a troubling sign that bind parameters with prepared statements aren’t being used in SQL statements which can make SQL injection attacks a possibility.

I re-read the thread and saw the tip about using “Always Enqueue”. That fixed the issue under the Enfold theme where the CSS wasn’t loading.

@nickyoung87
There does seem to be an incompatibility with the Enfold theme and this plugin. When I changed themes the calendar also resized better. I can’t change themes though and I haven’t had any problems with other plugins. Theme check comes up with lots of issues.

None of these CSS files get loaded under enfold theme:
/wp-content/plugins/google-calendar-events/assets/css/default-calendar-grid.min.css?ver=2abc3da3bd0a7cb81f41df90735aaad5

/wp-content/plugins/google-calendar-events/assets/css/vendor/qtip.min.css?ver=2abc3da3bd0a7cb81f41df90735aaad5

/wp-content/plugins/google-calendar-events/assets/css/default-calendar-list.min.css?ver=2abc3da3bd0a7cb81f41df90735aaad5

The navigation links don’t work on my site with the Enfold theme (in either grid or list). You can see both on this page.

https://www.stmatthews.us/calendar/

I will try checking for theme and plugin conflicts with a test site but I am debating whether that time wouldn’t be better spent switching to the legacy plugin and re-creating my feeds.

Here are samples from two failed login events. The second one is older and wordpress has since been updated to 4.2.0. I am not sure if there is an issue with username, maybe they were actually trying to login as {login}?

Anonymous user from 172.31.26.43 16 hours ago
Failed to login with username "Administrator" because no user with that username exists warning

Key		Value
id		11285
logger		SimpleUserLogger
level		warning
date		2015-04-26 22:36:10
message		Failed to login with username "{failed_login_username}" because no user with that username exists
type
initiator	web_user
occasionsID	fc19fe068b4a1e111ed26b846cc30560
subsequentOccasions	1
rep			1
repeated		2
occasionsIDType		fc19fe068b4a1e111ed26b846cc30560
context_message_key	user_unknown_login_failed
_server_http_x_forwarded_for_0	74.208.163.254
_server_remote_addr	172.31.26.43
_message_key		user_unknown_login_failed
failed_login_username	Administrator
server_http_user_agent

From last week (wordpress has since been updated to 4.2.0)

Anonymous user from 172.31.26.43 Apr 17, 2015 at 7:57
Failed to login with username "{login}" because no user with that username exists warning

Key		Value
id		11253
logger		SimpleUserLogger
level		warning
date		2015-04-17 07:57:54
message		Failed to login with username "{failed_login_username}" because no user with that username exists
type
initiator	web_user
occasionsID	fc19fe068b4a1e111ed26b846cc30560
subsequentOccasions	1
rep			1
repeated		26
occasionsIDType		fc19fe068b4a1e111ed26b846cc30560
context_message_key	user_unknown_login_failed
_server_http_x_forwarded_for_0	86.60.201.162
_server_remote_addr		172.31.26.43
_message_key		user_unknown_login_failed
server_http_user_agent	Mozilla/5.0 (Windows NT 5.1; rv:29.0) Gecko/20100101 Firefox/29.0
failed_login_username	{login}

Yes the real IP is there, as is the remote_addr which is one of the load balancers. Thanks for looking into it.

_server_http_x_forwarded_for_0 86.60.201.162
_server_remote_addr 172.31.26.43

The failed login name is showing up as:
failed_login_username {login}

I don’t know if they are actually trying to login as {login} or what exactly is going on there.

I think this was working at some point but it is no longer working. I see the load balancer IP addresses in the simple history list instead of the forwarded-for IP address.

I updated and logged in and now the remote address is showing both my IP address and the corporate proxy that I am currently going out of. I suppose the corporate proxy is adding the header and then the Amazon ELB is appending the IP of the corporate proxy that it sees.

This is what simple-history is logging:
_server_remote_addr 156.80.xxx.xxx, 128.229.xxx.xxx

The 156.80.xxx.xxx is my actual IP, and the 128.229.xxx.xxx is the IP address that the Amazon ELB would see (corporate outbound proxy).

Just be aware that there may be more than one IP address in the _server_remote_addr field. That is fine with me. I don’t know if a user could try to trick people by putting that header in at the browser but the last IP address in the list is sure to come from the load balancer in front of the wordpress site so it can be trusted (if you trust your load balancer). If you only want one IP address you could check for commas and take the last value in the comma delimited list.

Thanks.