infiniflow
[Question] I was wondering why when I upload a PDF document (which contains images), and click to analyze it, the process gets stuck and no error message appears?
Is there an existing issue for the same bug?
- [X] I have checked the existing issues.
Branch name
1
Commit ID
1
Other environment information
No response
Actual behavior
您好，请问为什么上传PDF文档后（PDF文档包含图片），点击解析，怎么一直卡住，也没有报错？
Expected behavior
No response
Steps to reproduce
1
Additional information
No response
My case is very strange, I used master branch build image, and ragflow server can parse documents. But i down the server and run again, the parser didn't work.
ps aux|grep task There should be a procedure named task_executor. Check Redis connection.
ps aux|grep task There should be a procedure named task_executor. Check Redis connection.
ps aux|grep task There should be a procedure named task_executor. Check Redis connection.
the task_executor procedure exist, but task running all the time.
same issue。
同样的问题，你解决了没？
不要批量解析，要一个一个解析，一个解析好了之后再点击下一个。我2000+文件基本把我搞半死。 感觉批量解析应该有bug，使用批量解析了好几天总是卡死。
same issue。
同样的问题，你解决了没？
不要批量解析，要一个一个解析，一个解析好了之后再点击下一个。我2000+文件基本把我搞半死。 感觉批量解析应该有bug，使用批量解析了好几天总是卡死。
我这边一个一个解析也是一样的问题；
same issue。
同样的问题，你解决了没？
不要批量解析，要一个一个解析，一个解析好了之后再点击下一个。我2000+文件基本把我搞半死。 感觉批量解析应该有bug，使用批量解析了好几天总是卡死。
我这边一个一个解析也是一样的问题；
这么多人遇到了同样的问题，本质上是个bug。
Embedding takes time with CPU.嵌入需要 CPU 时间。
This is actually a bug and has nothing to do with the CPU or GPU. Try to fix it PLZ @KevinHuSh
ragflow：v0.17同样碰到这个问题， 看task表实际任务已经完成，由于没有把task的process_msg同步给document表, 所以在页面看没有进度更新。 可能是这里异常时没有释放锁导致的。
def update_progress():
lock_value = str(uuid.uuid4())
redis_lock = RedisDistributedLock("update_progress", lock_value=lock_value, timeout=60)
logging.info(f"update_progress lock_value: {lock_value}")
while not stop_event.is_set():
try:
if redis_lock.acquire():
DocumentService.update_progress()
redis_lock.release()
stop_event.wait(6)
except Exception:
logging.exception("update_progress exception")
代码修改为
def update_progress():
lock_value = str(uuid.uuid4())
redis_lock = RedisDistributedLock("update_progress", lock_value=lock_value, timeout=60)
logging.info(f"update_progress lock_value: {lock_value}")
while not stop_event.is_set():
try:
if redis_lock.acquire():
DocumentService.update_progress()
redis_lock.release()
stop_event.wait(6)
except Exception:
logging.exception("update_progress exception")
if redis_lock.acquired:
redis_lock.release()
