Caleb
Forum Replies Created
-
I would always tend to preliminarily “trust” anything that reports an error. And attempt to fix based on that. 🙂
However, I do see a failure here in the Backup-plugin’s error checking process/handling, that personally I would fix QUICKLY if I was the developer.
The process.. Your local issue to fix:
The Backup ZIP is first created locally on your disk.. That’s where YOUR local Quota issue is met. You need to check your local hosting limitations against your userid’s/account’s overall disk-storage usage. Then either get allowance for a higher quota (or unlimited), or remove some of the accumulated stuff to get down under the quota. Clean your space.
Notice, that a userid’s quota counts EVERYTHING.. Which means that you can have automated backups working fine for years, and then suddenly because of uploaded stuff over time (such as image galleries and other wordpress files) you blow into the Quota limit, and backup’s WILL start failing. Even though you “backup to DropBox”, that does not mean that the initial backup processing writes directly to the cloud. The backup file is first created locally. Then afterwards saved in the cloud. So, if you cannot create a full-size local backup file to be uploaded, you are toast.
The process.. The plugin’s issue:
Backups need to be of HIGHEST TRUST and INTEGRITY. Main requirement for any backup software. If you cannot trust the content in a backup to be correct and restorable, then the result cannot even be called a backup. Backup software that cannot be trusted is worthless. That includes producing the correct errors and maintaining integrity when something out of the plugin’s control happens. Such as your Quota issue, which the backup plugin has no control over.
So in this case, The backup plugin passes on the ZIP software’s Quota error message (from inc/class-create-archive.php), code sleeps for 1 second, and then according to your log simply proceeds with the next steps of the “save a backup” process after the zip creation apparently actually failed half-way through. Which in your configuration includes copying the actually only half-made and BAD temporary ZIP archive to DropBox. As if it was a good backup file.
But a ZIP file creation that stopped in the middle because of space or quota issues is NOT A BACKUP file. It is practically nothing but a junk file. A partially made archive, that is missing unknown parts of the intended backup is worse than worthless and almost worse than nothing. Worse because the log indicating success and the (failed) archive files uploaded to DropBox gives the site admin a false sense of security..
“See all the nice backup files I have in DropBox..” Each file in reality worthless, because it is a half-completed archive that does not actually restore a site the way it should. (Backup Parts missing)
So.. in my opinion, you need to fix your Quota issue. That will allow your backup’s to be completed correctly without this quota error.
The plugin owner (Inpsyde) subsequently need to fix the plugin’s error checking and process to ensure that when archive creation fails because of space or quota issues, ALL further processing stops, and failure is reported to the Admin. That includes making sure that corrupted (half completed) archives are not saved away or uploaded to the cloud as if they were actually backups. Because they are not. A corrupted backup is not a backup.
disk-“space” is a limitation on a disk (file-system).
Disk-“quota” is an artificial limitation that CAN be imposed on a user (account), that limits how much space they are allowed to use. Preventing that account from eating up all the space on a file-system.
Hence, your disk(s) or filesystem(s) can show as having lots of free space, but still the user or account gets blocked by it’s quota, if one is imposed.
Since all your sites are probably running under the same account, they will all see any quota problem, as suddenly no more space can be allocated.
You probably have an accumulation of old backup files that eventually made your account run over it’s quota, if you have one. Check your account status, or from a Linux prompt use the “quota” command.
This issue is not related to the backup plugin other that it happens to be attempting to write a larger zip-file, and the issue must be fixed in your hosting account, as quota limits EVERYTHING you do. Everything that tries to write on the disk.
Causes WP or plugin updates to fail, logging to fail, backups to fail, …, …@elseyadcock, Yes, that is the solution..
I did that later on, after looking at the code. Worked fine.
My mistake that I forgot to update this FYI. Sorry.Maybe just adding a better error message or documentation pointer that people will understand would be a solution. 🙂
Or, now that there is a “record” in these forums. 🙂
You might want to read the sticky post op top..
I know the sticky states that for resource reasons they’ll probably usually look at this FREE forum on Wednesdays, but based on the last activity and response, that might have changed to Mondays.
Complaining that “no one is helping you” will likely have the opposite effect.
Especially since most people using the free version of the plugin might not be using AWS S3 (not free) as a backup destination. 🙂Forum: Plugins
In reply to: [Sociable] [Plugin: Sociable] Hunting for more.png in 'images/option1/32'Varunwordpress, Temporary solution only.
Every time a plugin is updated, the old plugin directory disappears and new stuff is loaded in.
So you would have to do this after every Sociable plugin update.
Root solution to the problem is to fix the plugin code at the source, and make the problem go away permanently.
Forum: Plugins
In reply to: [Sociable] [Plugin: Sociable] 4.3.1 – Help us grow informationInquiring minds wanted to know. So did I, so I went and checked.
The new version of Sociable now has code to offload your site configuration and other info (including your configured admin email address) and send it off to BlogPlay. Happens in two separate places/files in the code.
As stated in the plugin admin screen:
Check if you want help Sociable grow.
This information be used only to improve Sociable.
You accept send us your blog configuration and blog nameWhat they send off to sociablepixel.blogplay.com
$posts = array();
$posts[“blog_name”] = get_bloginfo();
$posts[“blog_url”] = get_bloginfo(‘wpurl’);
$posts[“admin_email”] = get_bloginfo(‘admin_email’);
$posts[“language”] = get_bloginfo(‘language’);
$posts[“version”] = get_bloginfo(‘version’);
$posts[“blog_config”] = $skyscraper_options;Your blog config, URL, sociable config, your email address.
The good thing for BlogPlay is that they actually give you a choice. The bad is that they never tell you that they take your email address away with the other information.
OFFLOADING PEOPLE’s email addresses without telling them is BOGUS.
I HATE THIS ONE.. My email addresses in blogs are special aliases used for only that, and I would NEVER sign up to have it offloaded without me knowing it, and potentially used for ANY random purpose..There is NO logical reason for anyone to offload email addresses from blog configurations other than wanting to start sending unsolicited email. Whether the addresses are kept internally or in the worst case shared/sold.
(I am NOT implying that BlogPlay would ever sell these addresses. But they are for sure not picking up email addresses without wanting to send unsolicited email to them at some point.)SI-contact-form got kicked off all my sites when they started doing this and advertising Spam from vCita suddenly started arriving to the magic tracker address they offloaded from my system, and Sociable just went the same way. Off my systems.
Fortunately I never upgraded to this new, magic version. Glad I checked before upgrading.
As I mentioned in the post right before this one (about catching code 400 responses), the one issue if you use CDNs in Super-Cache is that you must add the string bwp-minify to your Super-Cache do-not-touch filter. Otherwise Super-Cache will attempt to redirect the BWP-Minify calls to the CDN, which will not work.
Otherwise the two together seem to work fine.
Forum: Plugins
In reply to: [Better WordPress Minify] [Plugin: Better WordPress Minify] 400 Bad RequestFYI..
I just figured out what is going on..
But first, I looked at the two earlier reports about the message.
You probably know this already, but for the sake of this documentation, I think I should make it clear how the ‘400 bad request’ appear.The Bad request message shows up instead of minified CSS, which is why people complain that the CSS is broken. It is not there.
For example, when it calls to minimize some of the CSS files, when looking at the ‘CSS code’ supposedly produced by the BWP minify URL showing in the html, rather than minified CSS, it instead has this message as its sole content:
<h1>400 Bad Request</h1><p>Please see http://code.google.com/p/minify/wiki/Debugging.</p>
This message is from ‘min/builder/index.php’.
Solution in my case:
Checking around in other logs I found, that it was a conflict with the the cache plugin, in my case WP Super Cache, using a CDN.
Along the way, the cache plugin decides to manipulate the CSS/JS URLs calling on bwp-minify, and redirect them to the CDN, adding another host and directory prefix.. The ‘400 Bad Request’ was coming from ‘min’, because the directory structure “over there on the CDN” is not what it expects, now with a prefix added to the path.
Adding another filter string to the cache plugin, telling it to stay away from URLs that has bwp-minify in them fixed it. Now bwp-minify calls stay off the CDN.
Forum: Plugins
In reply to: [Better WordPress Minify] [Plugin: Better WordPress Minify] 400 Bad RequestI am seeing this problem still. Fresh downloaded plugin.
This report was originally done 6 months ago.Is there a version with a fix?
I sent this earlier on another thread with the same issue, but Re: Login, but I refreshed this one instead, since it actually has the right subject. 🙂
I am seeing this problem still. Fresh downloaded plugin.
This report was originally done 6 months ago.Is there a version with a fix?
Ran, you stated “We did not use and will not use these addresses to send unsolicited email.”.
That might in some worlds sound real, but the plugin did in fact use its “hooked-in” nature to schedule (at both 2:50 AM and 3:00 AM on my system), pick up, and send very specific email from what the content names as “vCita Inc, support@vcita.com 218 Main st, suite 363, Kirkland WA 98033, +1 855 824 8244”.
Bypassing normal spam protections to send us Spam (UCE – Unsolicited Commercial Email). I was targeted by multiples of these emails across addresses. So was yeahwow. I still have the emails, if you do not remember what they looked like.
Emails that were set up to look as if they were coming from our own email addresses, sent from our own local servers, but with Reply-To: pointing to support@vcita.com, and all links pointing to vCita. Advertising the selling points of vCita, such as “increasing our web-leads”, “collect payments”, and so on.
There was absolutely no difference in nature between these “direct pumped” emails from vCita and all the other Spam I catch and track every day, other than the fact that because it had been hooked into an installed WordPress plugin running on my system, it suddenly had the power to circumvent my layers of DKIM, SPF, and DNSBL blockers.
PLUS.. If one clicks on the link in one of the emails, one runs straight into a vCitas advertising landing page, with all of FSCF affiliate ID, the originating Blog domain (mine) AND the email address (mine) from one’s contact-page stuck in the URL to be nicely logged away as the source. Contact email addresses in my case created specifically for receiving contact email and nothing else. They would never otherwise have been passed on to any service, anywhere. Before this, they were fresh as newly blooming daisies.
So, “We did not use and will not use these addresses to send unsolicited email”??
That is certainly up for a debate. I know that I never “solicited” emails from vCita or voluntarily signed up for anything, and yet I got multiple “unsolicited” emails that are clearly advertising vCita and from vCita.
Hmm..
OK with me.. I will certainly “stop going on about it”, as requested, since apparently the creators think we are all just weird to care about “hidden features” and spamming.
To “stop caring”, I have just “fixed” the problem. I moved my contact pages to an alternate plugin, Contact Form 7. Works beautifully, has a ton of add-on plugins for extra functionality if needed, 6 Mill+ downloads, and it took me only about 5 minutes to swap into my pages.
So.. No need to fix anything from my perspective. I am out.
yeahwow,
To be fair. The issue I mentioned of the Banner link passing off email was not a new issue introduced.
The vCita ad banner was always setup so if you clicked on it, the URL would pass your email-address as a parameter to vCita. Most likely to pre-fill a sign-up form. For “user convenience” as Ran called it.
I think that in the hurry to fix some of the issues and get the 3.1.4.1 version out there, Mike probably just did not notice that additional issue, so it still showed in the new 3.1.4.1 version of the plugin.
BTW.. I think your idea of adding hooks in the plugin to allowing the install of a separate vCita add-on is a great idea. That would separate the issues entirely. A clean contact plugin, with a separate “Meeting” add-on people can choose to install.
Hmm.
“Sent from the plugin directly”.. Yes, exactly. That was the point.
But since the content and structure of those emails does not exist in the actual code, it would have been offloaded from outside.Every form set up by its default (in 3.1.4 at least) enables vCita and sends them the email contact entered in that form. By the time the form config is displayed and one turns off vCita, it seems to be too late already.
AND YIKES.. I just noticed that the Banner add that is shown at the top of the admin screen also sends ones email address off to vCita, if the vCita banner is randomly shown. (Since there are only a choice between a vCita ad or a ThemeFuse Ad, that would mean 50% of the time the admin screen is reloaded, the potentially changing email addresses in the various forms are shared with vCita.
BTW. The latter is still the case in the new 3.1.4.1 version.
I agree with you there.. Mike have a great plugin. Great work as I also stated in my other post. Until the connection with vCita.
I am personally OK with having vCita as meeting functionality (or better a service that does not take advantage of the connection by hooking themselves into my sites to use my system as a Spam sender).
I don’t mind Mike making some money from hooking up people that want such a service either. Good for him.BUT. It should be a non-default. No handing off random email-addresses to the meeting provider by default. If I were to sign up for a meeting service I would NEVER have used the email address I created for the contact form. They are assuming that this is an email address for public consumption, that I would want meeting requests on.
Also, it should be clearly documented up front what the side-effects will be both Spam and Privacy wise if one (as a personal, actual decision) enable the functionality in a plugin.
vCita’s method of abusing the connection seems very backhanded, when used on someone that never voluntarily signed up for their services or voluntarily gave them these multiple email addresses.
Why I would want my sites to start picking up Spam from someone else’s server in the middle of the night when am sleeping is beyond my comprehension. My server mail-logs clearly show that my web-server (apache user) started distributing Spam both at 2:50 AM and 3:00 AM, naming vCita as the culprit. Those timestamps matching my every 10 minute cron-job to call on wp-cron to refresh caches and assure scheduled postings.