Background:
I am getting a temperature float from an arduino via a serial connection. I need to be able to cache this temperature data every 30 seconds for other applications (e.g. web, thermostat controller) to access and not overload the serial connection.
Currently I cache this data to RAM as a file in /run (I'm trying to follow Linux convention). Then, other applications can poll the file for the temperature as they want it all day long with i/o now the only bottle neck (using an rpi, so not a lot of enterprise-level need here).
Problem:
I think when an app reads these files, it risks reading corrupt data. Should a writer update the file, and a reader try to read the file at the same time, can corrupt data be read, causing the thermostat to behave erratically?
Should I just use sqlite3 as an overkill solution, or use file locks (and does that risk something else not working perfectly)?
This is all taking place in multiple python processes. Is Linux able to handle this situation natively or do I need to apply somehow the principles mentioned here?
Calls to write(2) ought to be atomic under Linux.
Which means as long you are writing a single buffer, you can be certain that readers won't read an incomplete record. You might want to use os.write to make sure that no buffering/chunking happens you are not aware of.
if a read is happening and a file is updated, will the read use the new data while in the middle of a file, or does it somehow know how to get data from the old file (how)?
If there is exactly one read(2) and one write(2), you are guaranteed to see a consistent result. If you split your write into two, it might happen that you write the first part, read and then write the second part which would be an atomicity violation. In case you need to write multiple buffers, either combine them yourself or use writev(2).
Related
Though I would imagine that append mode is "smart" enough to only insert the new bytes being appended, I want to make absolutely sure that Python doesn't handle it by re-writing the entire file along with the new bytes.
I am attempting to keep a running backup of a program log, and it could reach several thousand records in a CSV format.
Python file operations are convenience wrappers over operating system file operations. The operating system either implements this file system operations internally, forwards them to a loadable module (plugin) or an external server (NFS,SMB). Most of the operating systems since very 1971 are capable to perform appending data to the existing file. At least all the ones that claim to be even remotely POSIX compliant.
The POSIX append mode simply opens the file for writing and moves the file pointer to the end of the file. This means that all the write operations will just write past the end of the file.
There might be a few exceptions to that, for example some routine might use low level system calls to move the file pointer backwards. Or the underlying file system might be not POSIX compliant and use some form of object transactional storage like AWS S3. But for any standard scenario I wouldn't worry about such cases.
However since you mentioned backup as your use case you need to be extra careful. Backups are not as easy as they seem on the surface. Things to worry about, various caches that might hold data in memory before if it is written to disk. What will happen if the power goes out just right after you appended new records. Also, what will happen if somebody starts several copies of your program?
And the last thing. Unless you are running on a 1980s 8bit computer a few thousand CSV lines is nothing to the modern hardware. Even if the files are loaded and written back you wouldn't notice any difference
I constantly need to process large amounts of heavy CSV files in order to modify their content (transform some fields, enrich others with extra data, etc). I went through a bunch of solutions already, all of them with ups and downs:
file system-based solutions are straightforward but messy, with files renamed with "-.wip" extensions etc, rather uggly and laborious to restore when something goes wrong
db-based solutions are cleaner but heavier and can be tiresome to deploy in our environment
whatever the solution, I always end up maintaining in-house solutions that are not really convincing.
Therefore I'm looking for a python-based solution allowing me to basically:
open an input file
process its input
write the processed input in an output file
all that in a transactional mode
"Transactional mode" guaranteeing that if anything goes wrong (something or someone trying to access the input or output file during processing, exception raising, eventually software or hardware crash), the transaction gets rolled back, meaning the input file remains untouched and the output file is erased.
Ideally, a file system-based solution would be the best, without any database in the way.
I already had a look at acidfile which is interesting but isn't exactly transactional.
I am in the process of writing a program and need some guidance. Essentially, I am trying to determine if a file has some marker or flag attached to it. Sort of like the attributes for a HTTP Header.
If such a marker exists, that file will be manipulated in some way (moved to another directory).
My question is:
Where exactly should I be storing this flag/marker? Do files have a system similar to HTTP Headers? I don't want to access or manipulate the contents of the file, just some kind of property of the file that can be edited without corrupting the actual file--and it must be rather universal among file types as my potential domain of file types is unbound. I have some experience with Web APIs so I am familiar with HTTP Headers and json. Does any similar system exist for local files in windows? I am especially interested in anyone who has professional/industry knowledge of common techniques that programmers use when trying to store 'meta data' in files in order to access them later. Or if anyone knows of where to point me, as I am unsure to what I should be researching.
For the record, I am going to write a program for Windows probably using Golang or Python. And the files I am going to manipulate will be potentially all common ones (.docx, .txt, .pdf, etc.)
Metadata you wish to add is best kept in a separate file or database for all files.
Or in another file with same name and different extension or prefix, that you can make hidden.
Relying on a file system is very tricky and your data will be bound by the restrictions and capabilities of the file system your file is stored on.
And, you cannot count on your data remaining intact as any application may wish to change these flags.
And some of those have very specific, clearly defined use, such as creation time, modification time, access time...
See, if you need only flagging the document, you may wish to use creation time, which will stay unchanged through out the live of this document (until is copied) to store your flags. :D
Very dirty business, unprofessional, unreliable and all that.
But it's a solution. Poor one, but exists.
I do not know that FAT32 or NTFS file systems support any extra bits for flagging except those already used by the OS.
Unixes EXT family FS's do support some extra bits. And even than you should be careful in case some other important application makes use of them for something.
Mac OS may support some metadata by itself, but I am not 100% sure.
On Windows, you have one more option to associate more data with a file, but I wouldn't use that as well.
Well, NTFS file system (FAT doesn't support that) has a feature called streams.
In essential, same file can have multiple data streams under itself. I.e. You have more than one file contents under same file node.
To be more clear. Same file contains two different files.
When you open the file normally only main stream is visible to the application. Applications must check whether the other streams are present and choose the one they want to follow.
So, you may choose to store metadata under the second stream of the file.
But, what if all streams are taken?
Even more, anti-virus programs may prevent you access to the metadata out of paranoya, or at least ask for a permission.
I don't know why MS included that option, probably for file duplication or something, but bad hackers made use of the fact that you can store some data, under existing regular file, that nobody is aware of.
Imagine a virus writing it's copy into another stream of one of programs already there.
All that is needed for it to start, instead of your old program next time you run it is a batch script added to task scheduler that flips two streams making the virus data the main one.
Nasty trick! So when this feature started to be abused, anti-virus software started restricting files with multiple streams, so it's like this feature doesn't exist.
If you want to add some metadata using OS's technology, use Windows registry,
but even that is unwise.
What to tell you?
Don't add metadata to files, organize a separate file, or index your data in special files with same name as the file you are refering to and in same folder.
If you are dealing with binary files like docx and pdf, you're best off storing the metadata in seperate files or in a sqlite file.
Metadata is usually stored seperate from files, in data structures called inodes (at least in Unix systems, Windows probably has something similar). But you probably don't want to get that deep into the rabbit hole.
If your goal is to query the system based on metadata, then it would be easier and more efficient to use something SQLite. Having the meta data in the file would mean that you would need to open the file, read it into memory from disk, and then check the meta data - i.e slower queries.
If you don't need to query based on metadata, then storing metadata in the file might make sense. It would reduce the dependencies in your application, but in order to access the contents of the file through Word or Adobe Reader, you'd need to strip the metadata before handing it off to the application. Not worth the hassle, usually
I have noticed that python always remembers where it finished writing in a file and continues from that point.
Is there a way to reset that so that if the files is edited by another program that removes certain text and ads another python will not fill the gaps with NULL when it does next write?
I have the file open in the parent and the threading children are writing to it. I used flush to ensure after write the data is physically written to the file, but that is only good to do that.
Is there another function I seem to miss that will make python append properly?
One safe thing, OS independent, and reliable is certainly to close the file, and open it again on writting.
If the performance hindrance due to that is unacceptable, you could try to use "seek" to move to the end of file before writing. I just did some naive testing in the interactive console, and indeed, using file.seek(0, os.SEEK_END) before writing worked.
Not that I think having two processes writing to the same file could be safe under most circumstances -- you will end up in race conditions of some sort doing this. One way around is to implement file-locks, so that one process just write to the file after acquiring the lock. Having this done in the right wya may be thought. So, ceck if your application wpould not be in better place using something to written data carefully built and hardened along the years to allow simultanous update by various processes, like an SQL engine (MySQL or Postgresql).
The title could have probably been put better, but anyway. I was wondering if there are any functions for writing to files that are like what the ACID properties are for databases. Reason is, I would like to make sure that the file writes I am doin won't mess up and corrupt the file if the power goes out.
Depending on what exactly you're doing with your files and the platform there are a couple options:
If you're serializing a blob from memory to disk repeatedly to maintain state (example: dhcp leases file),
if you're on a Posix system you can write your data to a temporary file and 'rename' the temporary file to your target. On Posix compliant systems this is guaranteed to be an atomic operation, shouldn't even matter if the filesystem is journaled or not. If you're on a Windows system, there's a native function named MoveFileTransacted that you might be able to utilize via bindings. But the key concept here is, the temporary file protects your data, if the system reboots the worst case is that your file contains the last good refresh of data. This option requires that you write the entire file out every time you want a change to be recorded. In the case of dhcp.leases file this isn't a big performance hit, larger files might prove to be more cumbersome.
If you're reading and writing bits of data constantly, sqlite3 is the way to go -- it supports atomic commits for groups of queries and has it's own internal journal. One thing to watch out for here is that atomic commits will be slower due to the overhead of locking the database, waiting for the data to flush, etc.
A couple other things to consider -- if your filesystem is mounted async, writes will appear to be complete because the write() returns, but it might not be flushed to disk yet. Rename protects you in this case, sqlite3 does as well.
If your filesystem is mounted async, it might be possible to write data and move it before the data is written. So if you're on a unix system it might be safest to mount sync. That's on the level of 'people might die if this fails' paranoia though. But if it's an embedded system and it dies 'I might lose my job if this fails' is also a good rationalization for the extra protection.
The ZODB is an ACID compliant database storage written in (mostly) python, so in a sense the answer is yes. But I can imagine this is a bit overkill :)
Either the OS has to provide this for you, or you'll need to implement your own ACID compliancy. For example, by defining 'records' in the file you write and, when opening/reading, verifying which records have been written (which may mean you need to throw away some non-fully written data). ZODB, for example, implements this by ending a record by writing the size of the record itself; if you can read this size and it matches, you know the record has been fully written.
And, of course, you always need to append records and not rewrite the entire file.
It looks to me that your main goal is to ensure the integrity of written files in case of power failures and system crashes. There a couple of things to be considered when doing this:
Ensure that your data is written to disk when you close a file. Even if you close it, some of the data may be in OS cache for several seconds waiting to be written to the disk. You can force writing to disk with f.flush(), followed with os.fsync(f.fileno()).
Don't modify existing data before you are certain that the updated data is safely on the disk. This part can be quite tricky (and OS/filesystem dependent).
Use file format that helps you to verify the integrity of data (e.g. use checksums).
Another alternative is to use sqlite3.
EDIT: Regarding my second point, I highly recommend this presentation: http://www.flamingspork.com/talks/2007/06/eat_my_data.odp. This also covers issues with "atomic rename".