{
  "id": 53353,
  "title": "A problem with running kernels on Kaggle",
  "url": "/competitions/talkingdata-adtracking-fraud-detection/discussion/53353",
  "author_name": "",
  "post_date": "2018-03-29T15:58:04.653305700Z",
  "votes": 14,
  "comment_count": 18,
  "views": 0,
  "content": "<p>Hi all. A couple of days ago, I discovered something strange when I tried running one of my kernels on Kaggle. When I hit the \"Commit and Run\" button I immediately got an error with no log file generated. I thought okay, probably, I had messed up somewhere in the code. But when I returned to the kernel's home page I noticed that it is still running despite the error. So, I let it finish its work and it did just fine. After that I was able to inspect the log file and even submitted the output file to the competition. So, I thought that it is probably just a glitch and did not expect it to ever happen again. The problem is that the glitch still persists and every time when I run a new kernel on Kaggle I must go through a similar process. It is a bit annoying because I cannot see the kernel's progress in real time, so I was wondering if any of you experienced a similar issue and if yes, then how did you deal with it? It would be helpful to see if it is just me or if other people straggling with this problem as well. Thank you in advance -- any help will be greatly appreciated!  </p>",
  "messages": [
    {
      "id": "305913",
      "postDate": "03/29/2018 15:58:04",
      "content": "<p>Hi all. A couple of days ago, I discovered something strange when I tried running one of my kernels on Kaggle. When I hit the \"Commit and Run\" button I immediately got an error with no log file generated. I thought okay, probably, I had messed up somewhere in the code. But when I returned to the kernel's home page I noticed that it is still running despite the error. So, I let it finish its work and it did just fine. After that I was able to inspect the log file and even submitted the output file to the competition. So, I thought that it is probably just a glitch and did not expect it to ever happen again. The problem is that the glitch still persists and every time when I run a new kernel on Kaggle I must go through a similar process. It is a bit annoying because I cannot see the kernel's progress in real time, so I was wondering if any of you experienced a similar issue and if yes, then how did you deal with it? It would be helpful to see if it is just me or if other people straggling with this problem as well. Thank you in advance -- any help will be greatly appreciated!  </p>",
      "rawMarkdown": "Hi all. A couple of days ago, I discovered something strange when I tried running one of my kernels on Kaggle. When I hit the \"Commit and Run\" button I immediately got an error with no log file generated. I thought okay, probably, I had messed up somewhere in the code. But when I returned to the kernel's home page I noticed that it is still running despite the error. So, I let it finish its work and it did just fine. After that I was able to inspect the log file and even submitted the output file to the competition. So, I thought that it is probably just a glitch and did not expect it to ever happen again. The problem is that the glitch still persists and every time when I run a new kernel on Kaggle I must go through a similar process. It is a bit annoying because I cannot see the kernel's progress in real time, so I was wondering if any of you experienced a similar issue and if yes, then how did you deal with it? It would be helpful to see if it is just me or if other people straggling with this problem as well. Thank you in advance -- any help will be greatly appreciated!",
      "votes": null
    },
    {
      "id": "305932",
      "postDate": "03/29/2018 16:31:55",
      "content": "<p>I've been having the same problem. I think, in the process of trying to fix something else about the new kernel editor, they have introduced a new bug (hopefully temporary).</p>",
      "rawMarkdown": "I've been having the same problem. I think, in the process of trying to fix something else about the new kernel editor, they have introduced a new bug (hopefully temporary).",
      "votes": null
    },
    {
      "id": "305934",
      "postDate": "03/29/2018 16:35:03",
      "content": "<p>Andy, thank you for your quick reply. It is very reassuring that it is not just my problem. I hope it will be fixed soon.</p>",
      "rawMarkdown": "Andy, thank you for your quick reply. It is very reassuring that it is not just my problem. I hope it will be fixed soon.",
      "votes": null
    },
    {
      "id": "305939",
      "postDate": "03/29/2018 16:40:04",
      "content": "<p>I am experiencing the same issue since this morning.</p>",
      "rawMarkdown": "I am experiencing the same issue since this morning.",
      "votes": null
    },
    {
      "id": "305962",
      "postDate": "03/29/2018 17:15:14",
      "content": "<p>Me too experiencing the same. I hope this is temporary. </p>",
      "rawMarkdown": "Me too experiencing the same. I hope this is temporary.",
      "votes": null
    },
    {
      "id": "306001",
      "postDate": "03/29/2018 18:16:08",
      "content": "<p>Hi, @Alexey. Thank you for reporting! Can you share a screen or copy+paste of the error you are seeing?</p>",
      "rawMarkdown": "Hi, @Alexey. Thank you for reporting! Can you share a screen or copy+paste of the error you are seeing?",
      "votes": null
    },
    {
      "id": "306007",
      "postDate": "03/29/2018 18:23:16",
      "content": "<p>Hi Sean, sure: </p>\n\n<p><a href=\"https://github.com/graf10a/Kaggle/blob/master/Error.jpg\">https://github.com/graf10a/Kaggle/blob/master/Error.jpg</a></p>\n\n<p>(This is taken from an old public kernel that I was able to run just fine a few days ago. Click \"Download\" to make it bigger.)</p>",
      "rawMarkdown": "Hi Sean, sure: \n\nhttps://github.com/graf10a/Kaggle/blob/master/Error.jpg\n\n(This is taken from an old public kernel that I was able to run just fine a few days ago. Click \"Download\" to make it bigger.)",
      "votes": null
    },
    {
      "id": "306535",
      "postDate": "03/30/2018 15:39:18",
      "content": "<p>Same here, but the only difference in my case is that it initiates another run in parallel which runs as normal but I cannot see the live log interface. So hitting the commit and run fails that specific run with the same error posted and in parallel runs another instance for the same kernel which runs as normal in the background.</p>",
      "rawMarkdown": "Same here, but the only difference in my case is that it initiates another run in parallel which runs as normal but I cannot see the live log interface. So hitting the commit and run fails that specific run with the same error posted and in parallel runs another instance for the same kernel which runs as normal in the background.",
      "votes": null
    },
    {
      "id": "306553",
      "postDate": "03/30/2018 16:20:06",
      "content": "<p>Yes, this is pretty much what I see. Thank you for sharing this information.</p>",
      "rawMarkdown": "Yes, this is pretty much what I see. Thank you for sharing this information.",
      "votes": null
    },
    {
      "id": "306816",
      "postDate": "03/31/2018 04:25:32",
      "content": "<p>I am having the same problem</p>",
      "rawMarkdown": "I am having the same problem",
      "votes": null
    },
    {
      "id": "307229",
      "postDate": "04/01/2018 06:34:05",
      "content": "<p>I am also facing same issue.</p>",
      "rawMarkdown": "I am also facing same issue.",
      "votes": null
    },
    {
      "id": "307452",
      "postDate": "04/01/2018 16:23:33",
      "content": "<p>Thank you very much to everyone who responded! So, it looks like I am not the only one who are running into this problem with Kaggle kernels. And now we know that the Kaggle technical support team is aware of this issue (see Sean's post). It would be nice to hear from them if/when they are planning to fix it. Personally, I find it very important because I am using the Kaggle kernels very extensively -- my home laptop has only 8 Gb of memory which is obviously not enough in this competition. </p>",
      "rawMarkdown": "Thank you very much to everyone who responded! So, it looks like I am not the only one who are running into this problem with Kaggle kernels. And now we know that the Kaggle technical support team is aware of this issue (see Sean's post). It would be nice to hear from them if/when they are planning to fix it. Personally, I find it very important because I am using the Kaggle kernels very extensively -- my home laptop has only 8 Gb of memory which is obviously not enough in this competition.",
      "votes": null
    },
    {
      "id": "307477",
      "postDate": "04/01/2018 18:13:42",
      "content": "<p>Same here Alexey, I am also dependent on Kaggle kernels due to RAM limit on my laptop. I hope this gets fixed soon. </p>\n\n<p>btw, this issue has also been <a href=\"https://www.kaggle.com/product-feedback/53553\">raised</a>  in product feedback section today. I think, this bug will get priority if more kernel users reports this issue. </p>",
      "rawMarkdown": "Same here Alexey, I am also dependent on Kaggle kernels due to RAM limit on my laptop. I hope this gets fixed soon. \n\nbtw, this issue has also been [raised][1]  in product feedback section today. I think, this bug will get priority if more kernel users reports this issue. \n\n  [1]: https://www.kaggle.com/product-feedback/53553",
      "votes": null
    },
    {
      "id": "307969",
      "postDate": "04/02/2018 18:25:38",
      "content": "<p>Same issue</p>",
      "rawMarkdown": "Same issue",
      "votes": null
    },
    {
      "id": "308625",
      "postDate": "04/03/2018 20:07:58",
      "content": "<p>I am too facing the same problem.</p>",
      "rawMarkdown": "I am too facing the same problem.",
      "votes": null
    },
    {
      "id": "308900",
      "postDate": "04/04/2018 09:39:25",
      "content": "<p>Same issue </p>",
      "rawMarkdown": "Same issue",
      "votes": null
    },
    {
      "id": "308913",
      "postDate": "04/04/2018 10:24:31",
      "content": "<p>I think yesterday was a great day since I could see the logs in real time while running scripts. But today the problem is back again ;-) Tried several browsers but no chance.</p>",
      "rawMarkdown": "I think yesterday was a great day since I could see the logs in real time while running scripts. But today the problem is back again ;-) Tried several browsers but no chance.",
      "votes": null
    },
    {
      "id": "310024",
      "postDate": "04/06/2018 12:28:50",
      "content": "<p>I am facing some issue with auto save in the script editor. May be there is something with my internet connection. IDK. But the interface certainly looks fishy sometimes!</p>",
      "rawMarkdown": "I am facing some issue with auto save in the script editor. May be there is something with my internet connection. IDK. But the interface certainly looks fishy sometimes!",
      "votes": null
    },
    {
      "id": "310552",
      "postDate": "04/07/2018 23:47:48",
      "content": "<p>Ok, it looks like the problem is gone -- I am able to run my kernels and observe their progress in real time as of this morning. Many thanks to whoever is responsible for fixing it!</p>",
      "rawMarkdown": "Ok, it looks like the problem is gone -- I am able to run my kernels and observe their progress in real time as of this morning. Many thanks to whoever is responsible for fixing it!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 305932,
      "author_name": "aharless",
      "author_url": "",
      "post_date": "03/29/2018 16:31:55",
      "content": "<p>I've been having the same problem. I think, in the process of trying to fix something else about the new kernel editor, they have introduced a new bug (hopefully temporary).</p>",
      "votes": null,
      "replies": [
        {
          "id": 305934,
          "author_name": "graf10a",
          "author_url": "",
          "post_date": "03/29/2018 16:35:03",
          "content": "<p>Andy, thank you for your quick reply. It is very reassuring that it is not just my problem. I hope it will be fixed soon.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 305939,
      "author_name": "sohaibomar",
      "author_url": "",
      "post_date": "03/29/2018 16:40:04",
      "content": "<p>I am experiencing the same issue since this morning.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 305962,
      "author_name": "pranav84",
      "author_url": "",
      "post_date": "03/29/2018 17:15:14",
      "content": "<p>Me too experiencing the same. I hope this is temporary. </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 306001,
      "author_name": "skbergam",
      "author_url": "",
      "post_date": "03/29/2018 18:16:08",
      "content": "<p>Hi, @Alexey. Thank you for reporting! Can you share a screen or copy+paste of the error you are seeing?</p>",
      "votes": null,
      "replies": [
        {
          "id": 306007,
          "author_name": "graf10a",
          "author_url": "",
          "post_date": "03/29/2018 18:23:16",
          "content": "<p>Hi Sean, sure: </p>\n\n<p><a href=\"https://github.com/graf10a/Kaggle/blob/master/Error.jpg\">https://github.com/graf10a/Kaggle/blob/master/Error.jpg</a></p>\n\n<p>(This is taken from an old public kernel that I was able to run just fine a few days ago. Click \"Download\" to make it bigger.)</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 306535,
      "author_name": "nuhsikander",
      "author_url": "",
      "post_date": "03/30/2018 15:39:18",
      "content": "<p>Same here, but the only difference in my case is that it initiates another run in parallel which runs as normal but I cannot see the live log interface. So hitting the commit and run fails that specific run with the same error posted and in parallel runs another instance for the same kernel which runs as normal in the background.</p>",
      "votes": null,
      "replies": [
        {
          "id": 306553,
          "author_name": "graf10a",
          "author_url": "",
          "post_date": "03/30/2018 16:20:06",
          "content": "<p>Yes, this is pretty much what I see. Thank you for sharing this information.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 306816,
      "author_name": "yinanzhu",
      "author_url": "",
      "post_date": "03/31/2018 04:25:32",
      "content": "<p>I am having the same problem</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 307229,
      "author_name": "codename007",
      "author_url": "",
      "post_date": "04/01/2018 06:34:05",
      "content": "<p>I am also facing same issue.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 307452,
      "author_name": "graf10a",
      "author_url": "",
      "post_date": "04/01/2018 16:23:33",
      "content": "<p>Thank you very much to everyone who responded! So, it looks like I am not the only one who are running into this problem with Kaggle kernels. And now we know that the Kaggle technical support team is aware of this issue (see Sean's post). It would be nice to hear from them if/when they are planning to fix it. Personally, I find it very important because I am using the Kaggle kernels very extensively -- my home laptop has only 8 Gb of memory which is obviously not enough in this competition. </p>",
      "votes": null,
      "replies": [
        {
          "id": 307477,
          "author_name": "pranav84",
          "author_url": "",
          "post_date": "04/01/2018 18:13:42",
          "content": "<p>Same here Alexey, I am also dependent on Kaggle kernels due to RAM limit on my laptop. I hope this gets fixed soon. </p>\n\n<p>btw, this issue has also been <a href=\"https://www.kaggle.com/product-feedback/53553\">raised</a>  in product feedback section today. I think, this bug will get priority if more kernel users reports this issue. </p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 307969,
      "author_name": "towelie",
      "author_url": "",
      "post_date": "04/02/2018 18:25:38",
      "content": "<p>Same issue</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 308625,
      "author_name": "robinreni",
      "author_url": "",
      "post_date": "04/03/2018 20:07:58",
      "content": "<p>I am too facing the same problem.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 308900,
      "author_name": "fkrasnov",
      "author_url": "",
      "post_date": "04/04/2018 09:39:25",
      "content": "<p>Same issue </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 308913,
      "author_name": "ogrellier",
      "author_url": "",
      "post_date": "04/04/2018 10:24:31",
      "content": "<p>I think yesterday was a great day since I could see the logs in real time while running scripts. But today the problem is back again ;-) Tried several browsers but no chance.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 310024,
      "author_name": "bhuvan93",
      "author_url": "",
      "post_date": "04/06/2018 12:28:50",
      "content": "<p>I am facing some issue with auto save in the script editor. May be there is something with my internet connection. IDK. But the interface certainly looks fishy sometimes!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 310552,
      "author_name": "graf10a",
      "author_url": "",
      "post_date": "04/07/2018 23:47:48",
      "content": "<p>Ok, it looks like the problem is gone -- I am able to run my kernels and observe their progress in real time as of this morning. Many thanks to whoever is responsible for fixing it!</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "305913": "Hi all. A couple of days ago, I discovered something strange when I tried running one of my kernels on Kaggle. When I hit the \"Commit and Run\" button I immediately got an error with no log file generated. I thought okay, probably, I had messed up somewhere in the code. But when I returned to the kernel's home page I noticed that it is still running despite the error. So, I let it finish its work and it did just fine. After that I was able to inspect the log file and even submitted the output file to the competition. So, I thought that it is probably just a glitch and did not expect it to ever happen again. The problem is that the glitch still persists and every time when I run a new kernel on Kaggle I must go through a similar process. It is a bit annoying because I cannot see the kernel's progress in real time, so I was wondering if any of you experienced a similar issue and if yes, then how did you deal with it? It would be helpful to see if it is just me or if other people straggling with this problem as well. Thank you in advance -- any help will be greatly appreciated!",
    "305932": "I've been having the same problem. I think, in the process of trying to fix something else about the new kernel editor, they have introduced a new bug (hopefully temporary).",
    "305934": "Andy, thank you for your quick reply. It is very reassuring that it is not just my problem. I hope it will be fixed soon.",
    "305939": "I am experiencing the same issue since this morning.",
    "305962": "Me too experiencing the same. I hope this is temporary.",
    "306001": "Hi, @Alexey. Thank you for reporting! Can you share a screen or copy+paste of the error you are seeing?",
    "306007": "Hi Sean, sure: \n\nhttps://github.com/graf10a/Kaggle/blob/master/Error.jpg\n\n(This is taken from an old public kernel that I was able to run just fine a few days ago. Click \"Download\" to make it bigger.)",
    "306535": "Same here, but the only difference in my case is that it initiates another run in parallel which runs as normal but I cannot see the live log interface. So hitting the commit and run fails that specific run with the same error posted and in parallel runs another instance for the same kernel which runs as normal in the background.",
    "306553": "Yes, this is pretty much what I see. Thank you for sharing this information.",
    "306816": "I am having the same problem",
    "307229": "I am also facing same issue.",
    "307452": "Thank you very much to everyone who responded! So, it looks like I am not the only one who are running into this problem with Kaggle kernels. And now we know that the Kaggle technical support team is aware of this issue (see Sean's post). It would be nice to hear from them if/when they are planning to fix it. Personally, I find it very important because I am using the Kaggle kernels very extensively -- my home laptop has only 8 Gb of memory which is obviously not enough in this competition.",
    "307477": "Same here Alexey, I am also dependent on Kaggle kernels due to RAM limit on my laptop. I hope this gets fixed soon. \n\nbtw, this issue has also been [raised][1]  in product feedback section today. I think, this bug will get priority if more kernel users reports this issue. \n\n  [1]: https://www.kaggle.com/product-feedback/53553",
    "307969": "Same issue",
    "308625": "I am too facing the same problem.",
    "308900": "Same issue",
    "308913": "I think yesterday was a great day since I could see the logs in real time while running scripts. But today the problem is back again ;-) Tried several browsers but no chance.",
    "310024": "I am facing some issue with auto save in the script editor. May be there is something with my internet connection. IDK. But the interface certainly looks fishy sometimes!",
    "310552": "Ok, it looks like the problem is gone -- I am able to run my kernels and observe their progress in real time as of this morning. Many thanks to whoever is responsible for fixing it!"
  },
  "source": "meta"
}