User Guide | Q&A | 用户指南 | 问答

[LWsmile]

linux：HTTPConnectionPool(host='192.168.0.24', port=6801): Max retries exceeded with url: /listprojects.json (Caused by NewConnectionError('<requests.packages.urllib3.connection.HTTPConnection object at 0x7f0a78b2d828>: Failed to establish a new connection: [Errno 111] Connection refused',))
windows：HTTPConnectionPool(host='localhost', port=6801): Max retries exceeded with url: /jobs (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x0000000004589CC0>: Failed to establish a new connection: [WinError 10061] 由于目标计算机积极拒绝，无法连接。',))
这个我应该怎么解决呢？

[LWsmile]

配置文件我应该怎么配置呢？能不能发生产环境一个模板呢？

[my8100]

Make sure that Scrapyd has been installed and started.
Then visit the Scrapyd server (in your case: http://192.168.0.24:6801) to check the connectivity.

[my8100]

配置文件我应该怎么配置呢？能不能发生产环境一个模板呢？

FYI

How to efficiently manage your distributed web scraping projects

如何简单高效地部署和监控分布式爬虫项目

[my8100]

Leave your comment here for help, in English if possible.
使用过程中如有疑问，请添加评论，尽量用英语。

[zhangshengchun]

CPU occupancy is too high when i refresh the statistical information，and the web interface has been waiting for a long time. i using scrapydweb to manage 25 scrapyd server nodes on five virtual machines in the same lan，how can I solve it?

[my8100]

@zhangshengchun Try the latest version v1.0.0 (which fixed issue 14) and remember to set ENABLE_CACHE = True.
If the problem remains, please new an issue with details.

[zhangshengchun]

@my8100 The problem of high CPU usage has been solved. thanks a lot!

[zhangshengchun]

@my8100 scrapyweb work very well, its a greate project！now i have a question, the finished job list in
dashboard getting longer and longer，how to remove finished job info and log in dashboard?

[my8100]

@zhangshengchun You have to manually restart the Scrapyd service. (Note that this operation would reset the running jobs at the same time since it keeps all job data in memory. ) That's why ScrapydWeb displays the finished jobs in the descending order.

[my8100]

@zhangshengchun Now you can set up the DASHBOARD_FINISHED_JOBS_LIMIT item in v1.1.0 to control the number of finished jobs displayed.

[KimFu2]

did you hava a docker deploy version?

[my8100]

Why not try it yourself and share with us?

[KimFu2]

Why not try it yourself and share with us?

ok.If there is, I don't have to do it. If not, I will do it and share it.

[my8100]

FYI

$ cat docker-entrypoint.sh 
#!/bin/bash
mkdir /code/logs
/usr/bin/nohup /usr/local/bin/logparser -t 10 -dir /code/logs > logparser.log 2>&1 &
#TO KEEP THE CONTAINER RUNNING
/usr/local/bin/scrapyd  > scrapyd.log

[1034467757]

Hello,I have some questions.How can I start more than one spider in a single scrapyd server without using timer task?In my project,I need to start lots of spiders at the same time(like batch start),or,I need to stop spiders at one scrapyd server.Thanks for your reply !

[my8100]

It's not supported yet as I don't think it's a common practice. You can use the Requests library in a Python script instead.

[1034467757]

It's not supported yet as I don't think it's a common practice. You can use the Requests library in a Python script instead.

OK,thanks!

[KimFu2]

Is SCRAPYDWEB a true distributed crawler?Use the same queue like Scrapy-redis, or the crawler is running independently?

[my8100]

ScrapydWeb is a web application for Scrapyd cluster management, not a crawler. Have you read the readme and these tutorials?

[kachacha]

I would suggest you set up the jitter option to add a random component to the execution time,
as SQLite is being used in the background.

Set it up to perform custom tasks randomly?

[my8100]

Execute a task by random delay of [-N, +N] secs, defaults to 0. (docs)
It would help avoid writing database concurrently.

[kachacha]

.（docs）

Thank you ，I‘ll try it。

[kachacha]

my SCRAPYD_SERVERS have not port ,is error "None of your SCRAPYD_SERVERS could be connected.";Thank you for your reply.

[my8100]

@kachacha
What do you mean? Can you post the full log?

[kachacha]

Fail to decode json from http://scrapyd********.local:80/logs/stats.json: Expecting value: line 2 column 1 (char 1)

[kachacha]

@kachacha
What do you mean? Can you post the full log?

This problem has been solved. It's the port problem.

[my8100]

Fail to decode json from http://scrapyd********.local:80/logs/stats.json: Expecting value: line 2 column 1 (char 1)

Post the full log and the content of stats.json.
Or rename stats.json to stats.json.bak and restart logparser.

[kachacha]

'pip install logparser' on host 'scrapyd1.****.local:80' and run command 'logparser' to show crawled_pages and scraped_items.

[kachacha]

[Jiyanggg]

I am running the service with docker.
For the persistence of timed tasks, (the scheduled task still exists when restarting)
I have mounted in the "/scrapydweb/data" directory, but it has no effect.
What should I do? Thank you

[my8100]

Check out the .db files in the data path on your localhost to confirm that it’s mounted successfully.
Besides, ScrapydWeb v1.3.0 now supports both MySQL and PostgreSQL backends, simply set up the DATABASE_URL option to store the data.

[yutiya]

[2019-08-30 10:35:00,050] INFO     in apscheduler.scheduler: Scheduler started
[2019-08-30 10:35:00,061] INFO     in scrapydweb.run: ScrapydWeb version: 1.4.0
[2019-08-30 10:35:00,062] INFO     in scrapydweb.run: Use 'scrapydweb -h' to get help
[2019-08-30 10:35:00,062] INFO     in scrapydweb.run: Main pid: 5854
[2019-08-30 10:35:00,062] DEBUG    in scrapydweb.run: Loading default settings from /Library/Python/2.7/site-packages/scrapydweb/default_settings.py
Traceback (most recent call last):
  File "/usr/local/bin/scrapydweb", line 11, in <module>
    load_entry_point('scrapydweb==1.4.0', 'console_scripts', 'scrapydweb')()
  File "/Library/Python/2.7/site-packages/scrapydweb/run.py", line 37, in main
    load_custom_settings(app.config)
  File "/Library/Python/2.7/site-packages/scrapydweb/run.py", line 124, in load_custom_settings
    print(u"{star}Overriding custom settings from {path}{star}".format(star=STAR, path=handle_slash(path)))
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe6 in position 21: ordinal not in range(128)

你好，想知道是哪里出错的呢？谢谢

[my8100]

Go to the user home directory (e.g. /home/yourusername) and try again.
It’s recommended to use Python 3 instead.

[yutiya]

谢谢。我用的python2.7。删除run.py 124行print(u"xxx") 这行改成print("xxx") 就可以运行了。很感谢

[my8100]

@foreverbeyoung
Try to install scrapydweb in virtualenv.

[panithan-b]

Hello, I am able to set scrapy and scrapydweb and get them running as a separate docker containers, but the pages and items are N/A, not a number as shown in your example at http://scrapydweb.herokuapp.com

The crawlers can be run properly without any errors shown.
Here are what I have tried:

1. In scrapyd.conf, I set items_dir=/spider/items
2. In scrapydweb_settings_v10.py, I set SHOW_SCRAPYD_ITEMS = True
3. I was using an example of JsonWriterPipeline in https://docs.scrapy.org/en/latest/topics/item-pipeline.html. Now I'm trying to use the Feed Exports from https://docs.scrapy.org/en/latest/topics/feed-exports.html

When run, I expected some items from customspider in /spider/items, but actually there is no file there

If possible, I would like to know how you set the items setting for scrapyd and scrapydweb on that site so that I would try modifying it to work in my implementation.

[my8100]

@panithan-b
Run scrapyd, scrapydweb, and logparser outside dockers to figure out how they cooperate first,
then update your dockers accordingly.

[panithan-b]

Hello, I've tried running scrapyd, scrapydweb, and logparser outside docker and got a log file like this, still not sure why most fields are still "N/A" or null
https://pastebin.com/WqRZYcAB

[my8100]

What’s the content in file “/spider/logs/tzspider/cosmenet/7dc32862e10311e9bb640242ac130002.log"

[panithan-b]

It was the blank file. But now I just know what happened. I set the LOG_LEVEL to ERROR in the scrapy config file for staging environment so it won't print anything except the ERROR level when it runs properly, so it prints nothing.

Now I set it to INFO and finally able to see the log content, the number of page crawled, and scraped items. :)

[Ray916]

Hello，how to make the send_text work, I want to use email to alert, but I dont understand what the code in the send_text page do, thank you

[stone0311]

hi, When I run the ninth crawler on each server, it shows waiting to run. How do I set it up to increase the number of crawlers that each server runs?looking forward to your answer.

[kachacha]

hi, When I run the ninth crawler on each server, it shows waiting to run. How do I set it up to increase the number of crawlers that each server runs?looking forward to your answer.

Set in the config of scrapyd. ...\scrapyd\default_scrapyd.conf

[stone0311]

嗨，当我在每台服务器上运行第九个搜寻器时，它显示等待运行。我该如何设置它以增加每个服务器运行的搜寻器的数量？期待您的回答。

在scrapyd的配置中设置。... \ scrapyd \ default_scrapyd.conf

thank you

[zaifei5]

Why does my timer always run twice?

[icella]

How display real ip in "servers" view page - [scrapyd servers], not '127.0.0.1:6800'?

Now i configured scrapyd_servers in a tuple like ('username', 'password', '127.0.0.1', '6801', 'group'), and the web page dispaly '127.0.0.1:6800', but i wanto real ip , so how to configure SCRAPYD_SERVERS. help pls~~~

[PShiYou]

Hello, can you record a full scrapydweb tutorial video? I followed the tutorials on the Internet and reported all kinds of mistakes

[488443059]

An error occurred while uploading the project str object has no attribute decode