Which file path and line number to show in log? - php

I have written a logging library that will add entries to a file or output stream.
When the entry is added to the log, it also includes context information such as the file and line number of the call that was made to add the entry.
My question is: Should I use the file path and line number of the call that added the entry, or the caller that triggered the event to log the message?
trigger.php:
<?php
$do->something(); #line 2
do.php
<?php
class Do
{
public function something ()
{
$this->log->add('Did something.', E_USER_NOTICE); # line 6
}
}
Right now, the message "Did something." is logged as having come from do.php line 6. Is it better to change the logging class to say that the message came from trigger.php line 2?
EDIT
Clarifying a few things.
The library includes the whole call stack in the log entry. So in more feature rich logging classes, you can see a lot more than what the example above entails.
The library does make use of logging levels and will filter out levels that aren't needed. The example above does not show that, but you can look at the library itself to know for sure.
Unfortunately I wasn't clear on how the message is being represented in this one scenario. This very minor scenario will only show one file path and one line. Assuming this could not be changed, which file path and line number would be logged?
Thank you for your help!

Also, what if you have three level instead of two ? What if you have four ? I recommend you to go with the conventional practice of keeping trace of the line where the actual log() call appears.
class client
{
public function get_some_data()
{
$data = $this->get_my_data();
if (empty($data)) {
$this->log->add('no data in sight');
}
// lots of processing
if ($something) {
$this->log->add('data was corrupted');
}
}
}
//...
$client->get_some_data();
If you only log the upper level call, you would only know that it happened in get_some_data() but not where exactly. You might think "Yes but I can find it using the exact error that was logged", but when you are digging logs to pinpoint a problem that is really not what you want to be doing.
If you have several call to this function and you want more informations about which one logged the line, what you can do is store a minimal stack trace alongside your log, see debug_backtrace(). It can be as minimal as
2011-12-25 17:02:37 ERROR "data was corrupted" at /path/client.php:6 (called from /path/somefile.php:57)
I would also strongly recommend that you add different logging level, because the logging you want in dev/debug situation is not the one you want in production.

Related

imap_mail_move correct implementation

I'm working with the PHP class IMAP and the function imap_mail_move have some conceptual problem (which I believe isn't PHP specific), when taking in consideration the available mailbox quota. The problem is that this function moves one message to another folder, flag the original message with a delete flag, and if you want to delete it, you call imap_expunge. Ok, so let's suppose I have one mail account with quota (available storage) of 15MB. And I receive one message with 12MB. When I try to imap_mail_move this message to another folder, I get quota limit exceeds, because the function tried to copy the message to the destination folder, exceeding the available limit. Is there any way to get around this problem? I tried to use the function imap_mail_copy, but looks like this function creates a 'mirror' of the original message in another folder, and not actually moves it. When using the imap_mail_copy function with the option: CP_MOVE (to delete the message of the current folder after copying), it gives me the exactly same error of imap_mail_move...
I have this code so far (not sure if this matters):
foreach($messagesUID as $messageUID)
{
moveMessageToTrash($imapCon, $messageUID);
}
function moveMessageToTrash($imapCon, $UID)
{
imap_mail_move($imapCon, $UID, "TRASH", CP_UID);
imap_expunge($imapCon);
}

Baffled: PHP Fatal error: Exception thrown without a stack frame in Unknown on line 0?

I have found that one common reason for the error is an exception being thrown from within an exception handler. I'm quite sure this doesn't happen in the application I'm trying to debug... But I've put all the initialization processing lines at the top of index.php in a try/catch.*
It can apparently also happen because some things cannot be serialized to be stored in a session. At most this application stores arrays into the session (quite a bit), but I'm confident that it doesn't store anything too out of the ordinary in it.
Someone commented that it happened to them because their primary key needed to be CHAR(32) instead of INT(11). The PK's in this app are all INTs.
Other suggestions are that it could be a problem with PHP 5.3.3 fixed in 5.3.6, full disk, and a need to typecast a SimpleXML value. We do happen to be running PHP 5.3.3, but upgrading would have to be a last resort in this case. It hasn't always been doing this.
UPDATE/NOTE: I actually can't reproduce the error myself, only see it happening in the logs, see below paragraph for where I believe the error is happening...
* From the error logs, it seems likely that at least one place it is happening is index.php. I am deducing this only because it is indicated in some entries by a referring URL. The try/catch code is currently only around the "top" initialization portion of the script, below that is mostly the HTML output. There is some PHP code in the output (pretty straightforward stuff though), so I may need to test that. Here is the catch part, which is not producing any output in the logs:
} catch (Exception $e) {
error_log(get_class($e)." thrown. Message: ".$e->getMessage(). " in " . $e->getFile() . " on line ".$e->getLine());
error_log('Exception trace stack: ' . print_r($e->getTrace(),1));
}
Would really appreciate any tips on this!
EDIT: PHP is running as an Apache module (Server API: Apache 2.0 Handler). I don't think there are any PHP accelerators in use, but it could just be that I don't know how to tell. None of the ones listed on Wikipedia are in phpinfo().
As far as I can tell the MPM is prefork. This is the first I'd ever looked into the MPM:
# ./httpd -l
Compiled in modules:
core.c
prefork.c
http_core.c
mod_so.c
The problem
In short you have a exception thrown somewhere, you have no idea where and up until now you could not reproduce the error: It only happens for some people, but not for you. You know that it happens for other people, because you see that in the error logs.
Reproduce the problem
Since you have already eliminated the common reasons you will need to reproduce the error. If you know which parameter will cause the error it should be easy to locate the error.
Most likely it is enough if you know all the POST/GET parameters.
If you can't reproduce with just these, you need to know additional request headers. Such as user agent, accept-encoding,...
If you still can't reproduce, then it becomes very difficult: The error may depend on a state (a session), the current time, the source ip address or the like.
The custom log method
Let's start simple: To get all parameters you can write in the very beginning of the affected php file something like:
file_put_contents("/path/to/some/custom_error_log", date()."\n".print_r(get_defined_vars(), true), FILE_APPEND | LOCK_EX);
Don't forget that the custom_error_log file must be writable to your php application. Then, when the error occurs in the error log, find the corresponding lines in your custom_error_log file. Hopefully there are not to many requests per second so that you can still identify the request. Maybe some additional parameters in the error log like source ip can help you identify the request (if your error log shows that).
From that data, reconstruct a request with the same POST/GET parameters.
The tcpdump method
The next option that is very simple as well, but requires you to have root-access on your target machine is to install tcpflow. Then create a folder, cd into that folder and simply execute (as root) tcpflow "port 80". The option (port 80) is a pcap filter expression. To see all you can do with that, see man pcap-filter. There is a lot what these filter expressions can do.
Now tcpflow will record all tcp connections on port 80, reconstruct the full data exchange by combining the packages belonging to one connection and dump this data to a file, creating two new files per connection, one for incoming data and one for outgoing data. Now find the files for a connection that caused an error, again based on the timestamp in your error log and by the last modified timestamp of the files. Then you get the full http request headers. You can now reconstruct the HTTP request completely, including setting the same accept-encoding, user-agent, etc. You can even pipe the request directly into netcat, replaying the exact request. Beware though that some arguments like a sessionid might be in your way. If php discovers that a session is expired you may just get a redirect to a login or something else that is unexpected. You may need to exchange things like the session id.
Mocking more things
If none of this helps and you can't reproduce the error on your machine, then you can try to mock everything that is hard to mock. For example the source ip adress. This might make some stunts necessary, but it is possible: You can connect to your server using ssh with the "-w" option, creating a tunnel interface. Then assign the offending ip adress to your own machine and set routes (route add host ) rules to use the tunnel for the specific ip. If you can cable the two computers directly together then you can even do it without the tunnel.
Don't foget to mock the session which should be esiest. You can read all session variables using the method with print_r(get_defined_vars()). Then you need to create a session with exactly the same variables.
Ask the user
Another option would be actually ask the user what he was doing. Maybe you can follow the same steps as he and can reproduce.
If none of this helps
If none of that helps... well... Then it gets seriously difficult. The IP-thing is already highly unlikely. It could be a GEO-IP library that causes the error on IPs from a specific region, but these are all rather unlikely things. If none of the above helped you to reproduce the problem, then you probably just did not find the correct request in all the data generated by the custom_log_file-call / tcpflow. Try to increase your chances by getting a more accurate timestamp. You can use microtime() in php as a replacement for date(). Check your webserver, if you can get something more accurate than seconds in your error log. Write your own implementation of "tail", that gives you a more accurate timestamp,... Reduce the load on the system, so that you don't have to choose from that much data (try another time of day, load of users to different servers,...)
circle the problem once you can reproduce
Now once you can reproduce it should be a walk in the park to find the actual cause. You can find the parameter that causes the error by trial and error or by comparing it to other requests that caused an error, too, looking for similarities. And then you can see what this parameter does, which libraries access it, etc. You can disable every component one by one that uses the parameter until you can't reproduce anymore. Then you got your component and can dive into the problem deeper.
Tell us what you found. I am curious ;-).
I had such an error, too. Found out that I returned a sql object in my session class (that was used by the session_handler) instead of returning nothing or at least not the sql object. First look into your _write and _read methods, if you too return some incorrect stuff.
Notice: ... Unknown on line 0 - How to find correct line, it's NOT "line 0"
I realize this question has already been answered, but I'll add this since it may help someone:
I managed to (unintentionally) produce errors without a stack frame from a function which used its own error handler to maintain control of execution while calling a potentially "dangerous" function, like this:
// Assume the function my_error_handler() has been defined to convert any
// PHP Errors, Warnings, or Notices into Exceptions.
function foo() {
// maintain control if danger() crashes outright:
set_error_handler('my_error_handler');
try {
// Do some stuff.
$r = danger();
} catch (Exception $e) {
$r = 'Bad Stuff, Man!';
}
restore error_handler();
return $r;
}
The "untraceable failure" would happen at the end of the program execution if the logic in "Do some stuff" returned from foo() directly, bypassing the call to restore_error_handler(). What I took away from the experience is this:
PHP maintains a stack of error handlers which gets deeper/taller with each call to set_error_handler().
Bad Stuff can happen if you push error handlers onto the stack and don't clean up after yourself before the program exits "normally".
This was a tough bug to isolate - I basically narrowed the problem down to the above function and then stared at it until my eyes bled.
So how would I have tracked this down, knowing what I know now? Since I don't know of any way to inspect the PHP error handler "stack" directly, I'm thinking it might make sense to use a Singleton object to encapsulate all set/restore operations for PHP error handlers. At least then it would be possible to inspect the state of the Singleton before exiting the program normally, and if "dangling" error handlers are detected to generate a sensible failure/warning message before PHP freaks out.
Instead of wrapping code in a try/catch block, what happens when you register an exception handler? Clearly your try/catch block is not catching the exception, thus resulting in the errors logged to Apache. By registering a handler, you can be sure any uncaught exception is handled.
Also, if you're using namespaces in your application, make sure you write \Exception in your catch block (or include the Exception class via a use statement).
This may be a little late but one issue I discovered when moving a site from a local to a remote server. I was using Concrete5 cms had developed my site locally(windows 8 in xampp) and then uploaded to a remote server running Cent 0S
Windows mysql by default is case insensitive and created a lower case database. Once this was uploaded to the remote server I received the "Exception thrown without a stack frame in Unknown on line 0?"
I then corrected the database tables case and my site started working again.
For us, this error was due to inadvertently serializing SimpleXML objects.
If you are using SimpleXML objects with 5.3.3, make sure you are are casting the node values to whatever you need (e.g. string) if you are serializing the values in the session.
Before:
$token = $response->Token->Value;
/* token saved in session, results in line 0 error */
After:
$token = (string) $response->Token->Value;
/* token saved in session, no error */
I had completely the same error. A very spacial case: if you connect an unnamed function (closure) hook to an object instance's hook point. After that you try to serialize this object.
I had the same error after filling the Illuminate Eloquent model's Fillable property incorrectly. Note the last 3 elements of the array, one is missing a coma.
protected $fillable = [
'budget',
'routestatus' ,
'userroutenumber'
'totalmovingseconds',
'totalidleseconds'
];
I had the same error, it appeared upgrading server from centos 5 to centos 6 and downgrading PHP from 5.4 to 5.3. Actual issue was PHP apc, not configured properly. Check your APC. I was using Symfony2, so you might find some help at Symfony Unable to allocate memory for pool
one simple way to produce this error is an old server with register_globals = On. then you only need two lines of code:
<?php
$_SESSION["my_var"] = "string";
$my_var = new MyClass(); //could be any class, i guess
?>
as soon as you reload this page once, you'll get the Exception thrown without a stack frame in Unknown on line 0 - error. seems like there is a conflict between the instance of the class and the (session) variable.
at least this is how i got this annoying error which is so hard to debug.
This problem occurred for me when I changed the namespace on a few Symfony bundles. Deleting the files in the the symfony cache directory fixed the issue.
Likely you have a corrupt/inconsistent table in the database. Try dumping the database. If you get a error that's the time. Repair that table and the issue should go away.
It is for this reason why clean install works. The clean install is just that clean.
mysqlcheck should work but if it does not show and issue still do above.

PHP - debug_backtrace() crashes - what can I do?

I looked at the messages I could find, but didn't find an answer that appears to help. I have a routine that calls debug_backtrace() and then cycles through the elements, outputting them in a nice format for my log. This works fine in most cases.
In some calls, the system just hangs. Assuming the following:
$curStack = debug_backtrace();
foreach ($curStack as $myStackLevel) {
$test = gettype($myStackLevel);
// code to write out contents of $myStackLevel
// Log that we are getting the next item
}
// Log that the foreach is done
In some cases, PHP just dies - I can see the log that it processed the current item, but it crashes on the foreach line (or at least before I can programatically see the contects of the next $myStackLevel.
In one message here, I saw something about recursive calls and using serialize, but I can't use serialize (I use PDO and had the message "You cannot serialize or unserialize PDO instances") So that's out.
The funny thing is that, if it crashes, the PHP error system still provides a complete dump of the stack.... (I found this out when I tried to use serialize).
Any help is appreciated....
Thanks,
JustMeToo
Are you trying to output the contents of the object elements? It's likely that you have a recursive object graph that causes the PHP process to either overflow the stack or take too long to run.

What's the point of having both include and require constructs in PHP?

I'm writing a PHP application for the first time (other than toys and exercises), and I'm at a loss to understand why PHP includes both an include and a require construct.
Before you write an answer explaining the differences between the two, let me first say that I do understand the difference - include produces a warning and keeps on going, and require produces a fatal error. My question is: when would you want to include, but not require a file?
Maybe it's a failure of imagination on my part, but there don't seem to be any files in my application that I don't want to scream about if they're not there. Oddly, this doesn't make me want to use require since it seems impossible to properly handle a failed require, so instead I use a helper function along the lines of this (warning: air code):
public static function include($filename) {
if (is_readable($filename)) {
if (!#include($filename)) {
throw new FileNotFoundException("File deleted after readable check");
}
} else {
throw new FileNotFoundException("File missing or unreadable");
}
}
I guess what I'm asking is:
What sorts of files would you really want to optionally include in an application?
Whether or not you want a file to be required or optional, why wouldn't you just handle that in a function like a modified version of the code I wrote above?
I'd use require for loading files essential for the app itself, i.e. require 'database_config.php', require 'core.php'. If this fails, you want it to fail as fast, hard and merciless as possible, since something is obviously wrong with your app installation. In this state it's not even guaranteed that it could handle a thrown exception properly, and require produces a very clear error message without any extra code.
include should be used for things like template files that you want to use when the app is already up and running and can handle its own errors gracefully.
Example:
include 'error_handler.php';
set_error_handler('error_handler');
/* something bad happens */
Warning: 'error_handler.php' not found!
Error: Specified error handler "error_handler" doesn't exist.
At some point you simply rely on basic files, even if it's just your error handler. You'd have to introduce extra code to handle a missing error handler gracefully, and even then the best you could do is output some error and die (unless you want to get into the game of catching errors gracefully even if your error handler doesn't exist). It's best to simply require 'error_handler.php'; you can include and custom handle everything after this if you want to.
What sorts of files would you really
want to optionally include in an
application?
I normally use include with static HTML blocks (header, footer, menu, search box, survey form...) and require with PHP functions or classes. It's unlikely that the footer block will disappear but, well, I suppose it's a nice visual clue when I look at the code.
Whether or not you want a file to be
required or optional, why wouldn't you
just handle that in a function like a
modified version of the code I wrote
above?
The # operator is said to be expensive and, anyway, I think that including an external file is a basic operation that should be covered by basic language. Your code also adds a new dependency: if you store it in a required block your optional blocks will no longer be 100% optional ;-)

PHP using exit()

I am using Cakephp but this is a MVC/php doubt
letting the view display the message
vs
echo 'Invalid Data'; exit;
I would like to know is there any pitfalls in the second case like memory leak etc.. Which one is better
EDIT
In case of a ajax call is exit good. and what about memory leak and other issues . Are all variables deallocated
You should use a custom ExceptionHandler (set_error_handler / set_exception_handler) and throw an Exception if you encounter any errors (CakePHP should already provide an ExceptionHandler). Make some space in your view and if the ExceptionHandler/ErrorHandler has a message, show it there to let the user know.
Your second code will just produce a blank page containing the little text. Every user will appreciate if you show the message inside your usual page layout instead of producing a blank page (which looks broken to most people).
The Cake tools to signal errors to the user are session messages and error views.
For "passive" actions like view actions, you should throw a 404 or similar, possibly more specialized error, e.g. if the requested model does not exist:
function view($id) {
$data = $this->Model->read(null, $id);
if (!$data) {
$this->cakeError('error404');
}
...
}
See Error Handling with CakePHP.
For any POST action, you should return the user to the view and display an error message using $this->Session->setFlash('Error!') and appropriate error messages for each invalid form field. That's the default behavior of baked views and controllers.
Terminating the whole script with exit makes for a miserable user experience.
In general, you should avoid exit. Exit is an abnormal termination, and programs should not terminate abnormally. Even if an error occurs, there are still many things that needs to be done - cleanup, logging, notifying the user etc. After all, your operating system doesn't reboot every time it cannot open a file.
performance-wise (AJAX cals)
Use exit().
user experience-wise (standard site nav)
Show the error in a proper formated page keeping the user within your site.

Categories